Skip to main content

Full text of "Brief View of the Caste System of the North-Western Provinces and Oudh, Together With an Examination of the Names and Figures Shown in the Census Report, 1882: Being an Attempt to Classify on a Functional Basis All the Main Castes of the United Provinces, and to Explain Their Gradations of Rank and the Process of Their Formation"

See other formats

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 
to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 
publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 

We also ask that you: 

+ Make non- commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liability can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at http : //books . qooqle . com/| 

Brief view of 
the caste 


Provinces ... 

John Collinson 
Nesfield, United 
Provinces of ... 

Digitized by 



Digitized by V^jOOQIC 

Digitized by 


Digitized by 


Digitized by 



or TH1 








Being an attempt to Classify on a Functional Baals all the Main Cartel of the United Provinces, and 
to Explain their Gradations of Bank and the Process of their Formation. 

38th FE3BBITABT, 1885. 



By JOHN a NESFIELD, M.A., Oion, 
Inspector of SehooU, Oudh. 





Digitized by VnOOQ IC 

Digitized by 

G oogle m 


Page 2, Una 1.— For "the paper * read " this paper." 

„ 12 „ 56.— For "section 120" read "slokal20." 

„ 15 „ 41.— 'For "caste — gradation 1 ' re*d " caste gradation." 

„ 24 „ 44. — For " father supplies the new jars " read "potter supplies the 

new jars." 

„ 30 „ 41 . — For "stages of arts" read "stages of art" 

„ 36 „ 3. — For " coasidered " read " considered." 

„ 37 „ 43.— For "demed" read "deemed." 

„ 62 „ 11.— For "tasma" read "tasm&d." 

„ 64 „ 54.— For " though multiform" read " thou multiform." 

,, 72 „ 38.— For "mystery of Brahmfc" read "mystery of Brahma." 

„ 88 „ 30. — For " statements appenned " read " statements appended." 

„ 9S „ 15.— For "oreation; of a" read "creation of a." 

„ 103 „ 3. — For " not bora within" read "not born within." 

,. 114 „ 52.— For " embodiments of these" read "embodiments of those." 


by Google 

Dinitizfid hv 






The caste area ... ... ... ... %% .1 

Objects of the present paper ... ... ... 2 

Special report on Tharus and Bogshas ... ... ... 3 

Classification of the people according to religions ... ... 4 


Fdrm VIII of census report not strictly confined to Hindus, as it professes 

to be ... .... k 

{ Arrangement of castes in oensus report merely alphabetical ... ... 6 

Foreign races not returned by caste ... ... ... 7 

1 Classification of castes adopted in this paper ... ... te< g 

Grounds of this classification ... ... o 

• This classification opposed to the modern Aryan theory ... ... 10 

j And to the old fourfold division into Brahman, Kshatriya, Vaisya, and Sudra ... 1 1 

I I — The casteless tribes ; general description ... ... ... 12 

Geographical arrangement of the casteless tribes ... ... 13 

Simplification of the tribe names given in census report ... .... 14 

II.— Castes allied to the hunting state ... ... ... ... 15 

Arrangement according to social precedence ... ... ... 16 

Hereditary propensity to crime explained ... ... ... 17 

Simplification of the caste names given in census report ... ... 18 

III. — Castes allied to the fishing state ... ... ... ... 19 

Description of the Meo caste ... ... ... ... 20 

Other fishing castes ... ... ... ... ... 21 

Simplification of the caste names given in census report ... ... 22 

IV.- — Castes allied to the pastoral state ... ... ... ... 23 

Functional etymology of the caste names ... ... ... 24 

Gradations of the pastoral castes ... ... ... ... 25 

Pastoral castes shown to be not of Scythian blood ... ... 26 

Simplification of the caste names given in census report ... ... 27 

V. — Agricultural castes ... ... ... ... ... 28 

Functional etymology of the caste names ... ... ... 29 

Gradations of the agricultural castes ... ... ... 30 

The Lodha and Bayar castes ... ... ... ... 31 

The Kandu, Kurmi, and Eachhi castes ... ... ... 32 

The Tamboli and Mali castes ... *.. ... ... 33 

The Taga and Bhuinhar castes ... ... ... ... 34 

Simplification of the caste names given in census report ... ... 35 

VI. — Landlord and warrior caste ... ... ... ... 36 

Union of landlord and military functions explained ... ... 37 

Modern Chattris proved to be mainly of non-Aryan blood ... ... 38 

Present occupations of the Chattri caste ... ... ... 39 

Simplification of caste names given in census report ... ... 40 

Gradual effacement of old distinctions of functions ... ... .41 

VII. — Artisan castes ... ... ... ... 42 

Distinguished into two great classes ... ... ... 43-44 

Fint class, those preceding the age of metallurgy ... ... 45 

Functional etymology of the caste names ... ... ... 46 

The Dharkar or Bansphor caste ... ... ... ... 47 

The Bari caste ... ... ... .. ... 48 

The Chamar caste ... ... ... ... ... 49 

Digitized by 




The Mochi caste ... ... ... ... ... 50 

The Dhuna caste ... ... ... ... ... 51 

The Kori caste ... ... ... ... ... 52 

The Teli caste ... ... ... ... ... 53 

The Kalwir caste .*, ... ... ... ,.. 54 

The Kumh&r caste ... ... ... ... ... 55 

The Lunia caste ... ... ... ... ... 56 

Status of the above castes in Hindu township ... ... ... 57 

Simplification of the caste names given in census report ... ... 58 

VIII.— Second class, artisan castes coeval with metallurgy ... • ... 59 

Social precedence attaching to these castes ... ... ... 60 

Functional etymology of the caste names ... ... ... 61 

The class of Sangtarash... ... ... ... ... 62 

The caste of Kamingar and Tirgar ... ... ... ... 63 

The Barhai caste ... ... ... ... ... 64 

The Gokain caste ... ... ... ... ... 65 

The Loh&r caste ... ... ... ... ... 66 

The castes of Kasera and Thathera... ... ... ... 67 

The Sonar caste ... ... ... ... ... 68 

The castes of Manih&r and Turkih&r ... ... ... 69 

The castes of Darai, Patwa, Chhipi, and Rangrez ... ... 70 

The Halwai caste ... ... ,. ¥ ... --# 71 

Simplification tf the caste names given in census report ... ... 72 

The Aryan theory in relation to the preoeding groups of castes ... 73 

IX.— The trading castes ... ... ... ... ... 74 

Sense in which the word "trading" is used in this paper ... ... 75 

Gradations of rank among the trading castes ... ... ... 7$ 

Arrangement of trading castes according to rank ... ... 77 

Multiplicity of trading castes explained ... ... ## 73 

Etymology of the caste names chiefly functional ... ... 79 

The Banj&ra caste ... ... ... ... _ g 

The Kunjra caste ... ... ... ... ### g| 

The Bhunja caste ... ... ... ... ^ 32 

The castes of Roniya, Kuta, Belwar, Bhartiya, and Lohiya ... ... 83 

Seven more trading castes ... ... ... ##> 34 

Social importance of banking castes explained ... ... ... 35 

The castes of Agrahari and Agarw&la ... ._ gg 

The Bohra caste ... ... ... ... ... 37 

The Khattri caste ... ... ... ... _ 33 

Prevalence of Jainism among trading castes ... ... ... 39 

Ethnic relations of upper trading castes ... ... ... 90 

Simplification of the caste names given in census report ... 91 

X.— The serving castes ... ... ... ... #- 92 

Functional etymology of the caste names ... ... a> 93 

The Bhangi caste ... ,.. ... ... 94 

The Dhobi caste ,.. ... ... ok 

The Kahar caste ... ... ... -## ^ 95 

The Napit or Nai caste ... ... ... ... 97 

The castes of Pawariya, Dhari, and Dom Mirisi ... ... 93 

-The Kathak caste ... ... ... ... ... 99 

The Bhat caste ... ... ... ... ... jOO 

The Eayasth caste ... ... ... ... ... jqi 

Extinction of the Baidya caste ... ... ... # jq2 

Simplification of the caste names given in census report ... ... 103 

XL— The priestly castes ... ... ... ... ... 104 

Numerical preponderance of Brahmans ... ... ... 105 

Sacrifice the function which called the Brahman caste into existence ... 106 

Digitized by 




Importance of the Brahman oaate arose from the extraordinary powers 
ascribed to sacrifice ... 

Technical rank of Brahmans in the orders of the Vedio priesthood 

Sense attached to the word " Brahman " in the Upanishads 

Subdivisions of Brahmans according to tribes, orders, clans, and branohes 

Subdivisions of Brahmans according to functions 

First function, Hotri 

Second function, Bidua 

Third function, Acharya 

Fourth funotion, Dikahit 

Fifth function, P&thak... 

Remaining functions less dignified than the preceding ones 

Sixth function, Jyotishi 

Seventh function, Pauranik 

Eighth function, Purohit 

Ninth function, Pande 

Tenth function, Ojha ... 

Eleventh function, Panda ... ... ... ,.. 

Twelfth function, Gangaputra 

Thirteenth function, Joshi ... ... ... ««, 

Fourteenth function, MahaVBrahman 

Extent to which the above functions have given rise to separate sub-castes, 

Incompleteness of statistics in census report ... 

One characteristic common to all the above functions 

Custom of feeding Brahmans explained ... ... •« 

To exercise no function at all is the highest dignity of a Brahman 

Probable proportion of Brahmans who exercise priestly functions 

' Occupations of Brahmans other than priestly ... 

Description of the Sunauriy a Brahman 

Brahmans are in the main not of Arykn blood ... 

XII. — The religious orders » . „ •«, 

Origin of the religious orders 
Ceremonies of induction 
Classification of the religious orders 
Shiva the type of the ascetic devotee 
Religious orders distinguished by badges ... .... 

Religious orders distinguished by austerities ... 
Religious orders founded on pantht or ways 
Goshains a caste as well as an order 
Causes which changed them into a caste 
Function of the Goshain caste as priests 
Occupations of Goshains other than priestly ... 
Bairagis threatening to become a caste 
Rivalries of the religious orders 
Incompleteness of statistics in census report 
Probable total number of religious devotees 

General summing up of results. 
Reduction of the number of Pastes shown in the census report ... 
Classification of oastes coincides with the progressive stages of industry 
Function the sole foundation of caste according to Hindus themselves 
This view confirmed by the analysis of caste names ... 
And by the industrial development of mankind in other countries 
Function the basis on which castes are still being formed 
Change of function the means by which individuals can change their caste ... 
Caste not based on a common ancestry or a common religion, bat a corn- 

mon function 
This is proved by the worship of functional tools, Ac. 












Digitized by 





And by that of functional deities or saints ... ... ... 162 

Names of castes not religious, but functional ... ... ... 163 

Sects have not given rise to castes ... ... ... ... 164 

Co-existence of several sects within a single caste ... ... 165 

Heredity of funotion the law of industrial development everywhere ... 166 

But this alone does not constitute caste ... ... 167-68 

Until it is supplemented by limitations of marriage ... ... 169 

Formation of the Brahman caste on this double basis ... ... 170 

Example of Brahman* followed by all the other functional olasses of the* 

community ... ... ... ... ... 171 

This is shown in the frequent recurrence of the number seven ... ... 172 

Explanation of Indian caste partly Brahmanical and partly industrial ... 173 

Monopoly of function prescribed in theory, but not possible in practice ... 174 

Evils of caste are social rather than industrial ... ... ... 175 

Tribe as distinguished from caste ... ... ... ... 175 

Diversity and not unity of funotion the characteristic of a tribe ... 177 

Caste has acted as the solvent of tribes ... ... ... 178 

Example taken from the Kol tribe ... ... ... ... 179 

Example taken from the Dom tribe ... ... ... 180 

Example taken from the Gaud tribe ... ... ... ... 181 

Same rule exhibited in the formation of a oaste out of other castes ... 182 

Example of the Khattri caste ... ... ... ... 183 

Example of the Bohra caste ... ... ... ... 184 

Example of the Kayasth caste ... ... ... ... 185 

Reason why certain partially-formed castes cannot become quite formed ... 186 
The organisation of the tribe was the model upon which the oaste has 

constructed itself ... ... ... ... .;. 187 

Firstly in regard to rules of marriage ... ... 1 88 

Secondly in regard to eating and drinking with outsiders ... ... 189 

Exceptional position of Brahmans in regard to eating and drinking ... 190 

Convivium follows connubium, but not vice lersd ... ... 191 

Probable number of castes in the province not less than 1,000 ... ... 192 

Changeableness of the marriage circle ... ... ... 193 

Numerical bulk of Chattris and Brahmans explained ... ... 194 

Final summing up of the theory of the Indian oaste ... ... 195 


Incompleteness of information in the census report ... ... 19$ 

Original tenets of Jainism were anti-Brahmanioal ... ... 197 

But have gradually assimilated to Brahmanism ... ... 193 

Castes exclusively Jain ... ... ... ... 199 

Function of the Jain castes ... ... ... ... 200 

Castes partly Jain and partly Hindu ... ... ... 201 

Keligious orders of Jains ... ... ... ... 202 

Jain ceremony of Rathjatra ... ... ... ... 203 


Caste alien to the spirit of Islam ... ... ... ... 204 

Twofold division of Indian Muhammadans ... ... 205 

First division, the four Muhammadan tribes ,.. ... ... 206 

These tribes are not castes ... ... ... ... 207 

But are to some extent marks of social status ... ... ... 208 

Second division, converts from Hindu castes ... ... ... 209 

High-caste converts and low-caste ones f#% .., ... 210 

Muhammmadan Rajputs ... ... ... ... 211 

In what respects they resemble Hindu Rajputs ... ... — 212 

Muhammadan Brahmans ... ... ... ... 213 

List of low-caste converts ... ... ... ... 214 

These are castes in every respect except the rule against eating and drink- 
ing with outsiders ... ... ... ... 215 

Digitized by 



They are considered to be castes in their own estimation 

Remarriage of widows common to both Hindus and Muhammadans 

Tendency of some castes to become exclusively Muhammadan 

Muhammadan religious orders 

Suggestion as to compiling statistics of religion 

Suggestion as to compiling statistics of occupation 

Incorrect identifications of certain caste names 

Incorrect entries of numbers against certain castes 










Digitized by 


Digitized by 









On the 17th February, 1881, a census was for the first time taken of the two pro- 
vinces (Oudh and the North- West) combined. Prior to this there had been three 
censuses of the North- West and one of Ondh. In the report of the census taken in 
1881 the results were published without any distinction of province; and as the inter- 
provincial landmarks were purely artificial, the statistics were set forth in a more 
compact form than they had ever been before. The tribes and castes of Oudh are 
identical with those of the North- West, by which they are on all sides surrounded. 
In fact, if the caste area could have been extended so as to include the Delhi districts 
of the Pan jab and the Behar districts of Bengal, the range of view would have been 
still more complete. For this is the area of Hindustan proper, and the caste system 
within this arfea has a distinct character of its own, differing in some respects both 
from that of the Indus valley to the west of it and from that of the Qangetic delta to 
the east known as Bengal proper. 

2. The dissertation which follows makes no pretensions to going beyond the 
area of the North-Western Provinces and Ondh, and it takes as its basis the names 
and figures collected in the census of 1881 and published in 1882 in the report com- 
piled by Mr. Edmund White, OS. It represents an attempt, which 1 was deputed to 
undertake nine months ago, in addition to my ordinary duties, to carry out the sug- 
gestion contained in the Home Department Circular No. 4 , dated 21st September,. 
1882, to the address of the Local Government, vit , "to collect statistics of castes and 
occupations and thus utilize the information brought together at the last census/' and 
also to make some proposal as to the shape in whioh the several castes might be 
arranged, with a view " to the statistics being eventually published in a uniform man- 
ner for each province." As it was presumed that the report compiled by Mr. White 
embodied all the information which the district officers had collected at the time of 
taking the census, I thought it better to examine the subject from a purely independ- 
ent standpoint —that is, in the light of such information as I happened to possess 
already respecting Indian castes or have been able to procure si&ce from private 
sources of various kinds. With the help of M. Ambika Prasad, Deputy Inspector of 
Schools, Lucknow, who was appointed to aid me in this work, and whose assistance 
has throughout been most valuable, enquiries have been addressed to various quarters, 
wherever there seemed to be any likelihood of useful information being procured. 
My aim has been to show to what extent, if any, the number of castes enumerated in 
the census report can be reduced and simplified, and what castes, if any, have been 
omitted ; to describe in general terms the occupation, both past and present, for which 
each tribe and caste is distinguished— a point about which nothing is said in Section 
XXIII on " castes" in the preliminary dissertation of the report ; and to devise some 
plan for arranging and classifying the several tribes and castes, which in form VIII 
of the census report are given merely in alphabetical order. 

3.. A minute description of the industries, customs, and traditions of every single* 
tribe or ca3te would, of course, be a work of many years. Nothing more, therefor e> 


Digitized by 


( 2 ) 

than a general account of each group has been attempted in the paper. But as a speci- 
men of the enquiries which are being made respecting individual tribes or castes, and 
of the nature of the results which might be elicited if these enquiries are continued, 
I attach herewith an account of the Th&rus and Bogshas, containing as minute a 
description as I could compile out of the information collected by myself since June, 
1884, when this work was first commenced. 

4. In form I of the twenty-one census forms prescribed by the Census Commis- 
sioner for India, the total population of the North- Western Provinces and Oudh, in- 
cluding persons of all creeds and nationalities, is shown as 44,107,869. In form III the 
total population, classified according to religions, is shown as follows :— 


Hindus ... 






Sikhs ... 








Jews — 


















Total .< 

If we exolude Nob. 4-9, which have nothing to do with the subject of Indian caste, 
and attend only to Hindus, Jains, and Muhammadans, the remainder is 44,056,227. 


5. To take the Hindus first. The total just shown (38,053,394) tallies with the 
total given at the bottom of form VI II, and this form professes to give " details of 
Hindu castes for the province" (vide table of contents)* It is clear, therefore, that 
none but Hindu castes are intended to be included in this form. But the intention 
does not appear to have been consistently acted out. For some of the castes included 
in the statement are not Hindu at all, but Muhammadan. This is true, for example of 
the Jul&ha or Muhammadan weaver caste, the Ghosi or Muhammadan milkman caste, 
and of the Gharkata or Muhammadan caste of elephant-keeper. It is also to a large 
extent true of fakirs or the Muhammadan mendicant classes, and of Kunjras or the 
caste of green-grocers, Kam&ngars or the caste of bowmaker and painter, P&warias 
or the caste of musicians, and many more. 

6. The detailed list of Hindu castes given in form VIIIB consists of three 
different parts : (a) a list of one hundred and eighty castes given in alphabetical order, 
(b) alist of eleven " races not returned by caste" and given, like the preceding, in alpha- 
betical order; (<?)an aggregate of 16,121 persons under the single heading of "un- 

7. About the " unspecified " there is of course nothing further to be said. As 
to the list of " races not returned by caste, " it appears that the principle has not been 
strictly carried out ; for in the preceding list of 180 castes some names are entered 
which refer to races or tribes and not to castes, as, for example, Gurkhas, Tharus, 
Saharias, and others. Perhaps the list was intended for foreign races or tribes — that 
is, persons or communities not indigenous to the province, but immigrants from other 
provinces or territories who have become domiciled within its area, but have not 
amalgamated with the native population, so as to form a real part of it ; such persons, 
for example, as Bengalis, Gujaratis, Kashmiris, &c. But if this was the principle 
intended (and such appears to have been the case), three tribes should have been 
omitted— namely, Bhil, Gond, and Eol, all of whom are to be found in the districts 
south of the Ganges and Jumna and are strictly indigenous to the soil ; and one more 
should have been added, viz., Gurkha, which is placed in the census report among 
the indigenous castes, though it is really a tribe of Nepal. 

Digitized by 


3 ) 

8. As the caste-names shown in the census report have been given in merely 
alphabetical order, the first thing to be done was to attempt to lay down some prin- 
ciple on which castes could be classified. The scheme which I have prepared is 
shown in detail in the thirteen tabular statements hereto appended. A brief summary 
of their contents may, however, be shown in this place in the following form : — 





Name of group. 


Number of 

and castes. 

Number of per- 
sons shown in 


Casteless tribes ... ... ... ... 

Castes connected with land— 

4. Allied to hunting: ettte ... ... ... 

B. Allied to fishing state ... .,. .,. 

C. Allied to pastoral state ... «. ... 

D. Agricultural. ... ... ... 

E. Landlords and warriors ... ... ~. 


Artisan castes — 

A. Preceding metallurgy ... ... — 

B. Coefal with metallurgy ... 

Totsl m( ... 

Trading castes — ... „. M . 

Serving castes ... M , 

Priestly castes 

Religious orders ... ... 

Total of the provincial tribes and castes 

Foreign races not returned by caste 

Unspeciied ... ••• ... ... 

Gband Total 










* 25 





















9* It will be seen that I havejdivided the native Hindn population of the pro- 
vince, as distinct from the foreign and the unspecified, into seven main groups or classes* 
The first consists of those backward and semi-savage tribes which have not yet been 
absorbed into caste, and which I have, therefore, designated by the name of the caste- 
lets tribes. The other six groups consists of the great divisions of caste into which 
the people of Hindustan have become gradually distributed and absorbed in the 
course of thousands of years, each group being distinguished from every other by 
some speciality of function which marks its general character. The two largest of 
these six groups (the group of landed castes and that of artisan castes) are further 
aub-divided into smaller groups, each representing some phase of industry peculiar 
to itself, the rank of which in the scale of human progress has determined the rank 
held by the castes corresponding to it in the Hindu social scale. The plan, therefore, 
on which the castes have been classified in this paper is intended to be something 
more than a merely symmetrical arrangement. It is intended to explain the order 
in which the several castes came into existence, and thus to account for the degree 
of social importance and respectability in which they stand to the general commu- 
nity. If, for example, the stage of culture is a low one, it will be invariably found 
that the caste corresponding to this stage and practising one or more of the industries 
peculiar to it stands low in the eyes of the Hindu community generally, and so on 
through all variations of caste and all stages of industry up to the highest. 

10. It must be here remarked in passing that sueh a theory as the above is not 
compatible with the modern doctrine which divides the population of India- into Aryan 
and Aboriginal. It presupposes an uubroken continuity in the national life from one 
stage of culture to another, analogous to what has taken place in every other country 
in the world whose inhabitants have emerged from the savage state. It assumes, 
therefore, as its necessary basis, the unity of the Indian race. While it does not deny 
that a race of " white-complexioned foreigners, 9 ' who called themselves by the name of 
Arya, invaded the Indus valley via Kabul and Kashmir some four thousand years ago 
and imposed their language and religion on the indigenous races by whom they found 

Digitized by 


(4 ) 

themselves surrounded, it nevertheless maintains that the blood imported by this 
foreign race became gradually absorbed into the indigenous, the less yielding to the 
greater, so that almost all traces of the conquering race eventually disappeared',, 
just as the Lombard became absorbed into the Italian, the Frank into the Gaul,, 
the Roman (of Roumania) intothe Slav, the Greek (of Alexandria) into the Egyptian, 
the Norman into the Frenchman, the Moor (of Spain) into the Spaniard ; and as the 
Norwegians, Germans, &c, are at this day becoming absorbed into Englishmen in 
North America, or as the Portuguese (of India) have already become absorbed into 
Indians. I hold that for the last three thousand years at least no real difference of blood 
between Aryan and aboriginal (except perhaps in a few isolated tracts, such as Raj- 
pntana, where special causes may have occurred to prevent the complete amalgama- 
tion of race) has existed ; and the physiological resemblance observable between the 
various classes of the population, from the highest to the lowest, is an irrefragable 
proof that no clearly-defined racial distinction has survived — a kind of evidence which 
ought to carry much greater weight than that of language, on which so many fanci- 
ful theories of ethno logy have been lately founded. Language is no test of race ; 
and the question of caste is not one of race at all, but of culture. Nothing has tended 
to complicate the subject of caste so much as this intrusion of a philological theory, 
which within ita own province is one of the most interesting discoveries of modern 
times, into a field of enquiry with which it has no connection. The "Aryan bro- 
ther" is indeed a much more mythical being than R&ma, or Krishna, or any other of 
the popular heroes of Hindu tradition whom writers of the Aryan school have vainly 
striven to attenuate into solar myths. The amalgamation of the two races (the Aryan 
and the Indian) had been oompleted in the Panjab (as we may gather from the Insti- 
tutes of Manu) before the Hindu, who is the result of this amalgamation, began to 
extend his influence into the Ganges valley, where by slow and sure degrees he disse- 
minated among the indigenous races those social and religious maxims which have 
been spreading wider and wider ever since throughout the continent of India, absorb- 
ing one after another, and to some extent civilizing, every indigenous race with 
whom they are brought into contact, raking the choice spirits of the various tribes 
into the rank of Brahman or Chhatri, and leaving the rest to rise or fall in the social 
scale according to their capacities and opportunities* 

1 1 . Equally incompatible with the theory advanced in this paper is that of the old 
semi-mythical division of caste into Brahman, Eshatriya, Yaisya, and Sudra. This classi- 
fication has. done almost as much to prevent men from studying and describing Indian 
castes as they are as the modern fiction which divides them into Aryan, aboriginal 
and mixed. But of the two,, the older doctrine comes much nearer to the truth, for 
it makes the differences of caste depend upon differences of functions, the only basis 
on which Indian castes have been formed j while the other, ignoring function or con- 
signing it only to the second place, attempts to make caste a question of blood and 
appeals to the futile evidence of language in support of this. It has been proved 
(conclusively as I think) by modern scholars that the old fourfold division was not 
even of Indian origin, and was never actually in force in India except as a current 
tradition, the only reality which attaches to. it to this day. Professor Kern and the 
late Dr. Haug, of the Educational Department, Bombay^ have independently shown that 
the same division of classes existed among the common ancestors of the Perso-Aryans 
and the Indo-Aryans, before the latter had separated themselves from their kinsfolk and 
established settlements in the Indus valley. In the Ydsna (XIX., 46) these four class- 
es are mentioned under tho names of Athrava, Rathaesthao, Vasttiya-fshuyant, and 
Huiti. The first Atharva is merely the Perso-Aryan original or equivalent of the 
Indian Atharvan or fire-priest, one of the numerous names for professional priest 
which, ifa common with " Brahman,'* are to be found in the Vedic hymns and liturgies. 
One of the Vedas, as is well known, is called the Atharva, and this was pre-eminently 
the manual of the " Brahman" or superintendent priest The second name, Rathaesthao 
merely signifies " one who rides in a chariot," and is therefore a correct equivalent to, 

Digitized by 


( 5 ) 

the Sanskrit " Ratheshtha," " Rajanya," and " Kshatriya," the names by which knight* 
and high-class warriors are called in the Vedic hymns. The third, Vastriya, comes very 
near in sonnd to the Indian " Vaisya;" and such resemblance is not at all irrecon- 
oileable with the fact that Indian grammarians have derived the word " Vaisya" from 
the root vi$h, " to inhabit" The fourth class, Haiti, answers both in name and function 
to the Indian-" Sudra;" for in the ancient society of Persia the Haitis were " workmen, 
servants, and petty traders/' and there is no known Sanskrit root to which " Sudra " 
can be traced. Professor Kern also quotes the following verse from Yaena, XIX, 
44 :— "This command, which the creative spirit has spoken, embraces the four classes 
(pi$htra\" that is, is intended to apply to the entire population, and he contends that 
pishtra is the equivalent of the Sanskrit varna. Nothing could be more natural than 
that the Aryan tribes, who migrated into India, should have preserved the tradition 
of the four classe s into which their own population was roughly divided, and that the 
names themselves should have gradaally acquired new applications and uses under 
the altered circumstances of society and country. After a modus vivendi had been 
established between the invading foreigner and the conquered indigenes, and the two 
races had begun to be fused, as they eventually were, into a single whole, the titles 
" Brahman-'" and u Rshattriya " were reserved to the priests and nobles of whatever 
blood, who in course of time became stereotyped into distinct castes ; while the name 
"Sudra" was applied to the lowest or menial classes of the people, and "Vaisya* to 
the intermediate ones. But there is nothing to show that the last two ever became 
castes as the first two did, and the written evidence that we possess is entirely against 
the supposition. These classes (Vaisya and Sudra) are very rarely, if ever, alluded 
to in the great national epics, the R&m&yan and Mah&bhirMa, the best authority that 
we possess for the state of society in ancient India ; whereas we are constantly remin- 
ded of the existence of Brahmaus and Kshatriyas, and of lower classes of men, like 
barbers, carpenters, fishermen, potters, and the like. Again in the M&nava Code 
the vagueness that hangs about the names of Vaisya and Sudra is in such striking con- 
trast with the clear and precise account given by the author respecting the Brahman 
and Kshatriya castes and the various industrial classes existing in his own time, that 
we are driven to suppose that such castes as Vaisya and Sudra did not really exist, but 
that these words merely expressed a rough classification of the people outside the radius 
of Brahmans and Kshatriyas. There seems, then, to be no reason for supposing that 
the old fourfold division of caste was indigenous to Indian soil, or that it ever really 
represented the caste system of the Indian people. In any case the caste system of 
the present day is something entirely different, and the attempts that have been made 
to adapt existing castes to the classifying framework of Vaisya and Sudra have 
resulted in nothing but confusion and failure* 

12. The easteleee tribes. — The list of tribes whom I have put together under the 
general designation of "the casteless tribes" will be seen in tabular statement, 
group I. These tribes are the last remains and sole surviving representatives of the 
aboriginal Indian savage, who was once the only inhabitant of the Indian continent, 
and from whose stock the entire caste system, from the sweeper to the priest, was 
fashioned by the slow growth of centuries. They enable us to form some faint idea of 
what the ancient tribes of Hindustan were like, say some six thousand years ago, before 
cattle-grazers, husbandmen, artisans, traders, and the priestly and royal clans had 
been differentiated into distinct social and industrial types. Their manners and 
notions will be found to correspond in many respects with those of the most backward 
races in various parts of the world, and it is only by looking at them from this 
comparative standpoint that their manners are worth studying. In religion these 
tribes cannot be called either Hindu or Muhammadan, though they have picked up 
some of the tag ends of both creeds through the inevitable operation of many centu- 
ries of contact. If the tools and weapons which they use are better than what could 
be found among the savages of the Pacific or in the continent of Africa, this is not 
because they themselves have advanced beyond the Stone age of the world, but only 


Digitized by 


( « > 

because they can readily supply themselves with iron implements and weapons 
manufactured by their more advanced kinsfolk in the Indian towns and villages amon g 
which they wander. The distinctive characteristic of almost all these tribes is that 
they lead a wandering life, are incapable of sustained toil^ and subsist to a large 
extent on the animals, roots> and fruits spontaneously provided by nature. There is 
no space in a paper like this, which aims at giving a rough description end analysis 
of the entire system of Indian castes, to furnish a detailed account of any one of these 
primitive tribes The essay on.Tharus and Bogsbas, referred to already in para. 3 
and published in the January number of the Calcutta Review, may serve as a 
specimen of what can. be done in this direction, if minute inquiries are made and 
leisure can be found for putting the results into shape. 

13, If the reader will refer to the list of names given in group I as shown in 
the annexed table, he will see that they have been arranged as far as possible in 
geographical order. The Th&rus and Bogshas, who head the list,, inhabit the sub- 
Him&layan forest which forms the northern boundary of Hindustan. The four tribes 
who are placed at the bottom of the list (see serial Nos. 15-19) dwell in the forests 
to the south of the Ganges and Jumna. The Bhils, (Jonds, and Saharias> all of whom 
are to be found in the Jh&nsi Division, belong to the Central Indian plateau rather 
than to the plains of Hindustan. The Kola however, whose name is mentioned last 
of all, belong strictly to the Ganges valley and are to. be found in considerable 
numbers in the forests of Mir*apur, Allahabad, and B&nda. A. peculiar interest 
attaches to this tribe, as being the savage ancestors of the great weaver caste called 
Koli or Kori, and of the great agricultural caste called Kori or Kachhi. The Thftrus 
in the north and the Kols in the south have retained many of tte simple virtues of 
the untutored savage! and to such people the forest is the natural home. The 
remaining tribes (omitting the Bhils, Gonds, and Saharias) dwell in the intervening 
deforested plain, and lead a precarious life as thieves, gang-robbers, jackal-hunters, 
trappers, fowlers, jugglers, acrobats, and hereditary prostitutes — savages whom the 
surrounding civilization has demoralized rather than improved, because it has failed 
to assimilate,, and whom Mann described as " sinful and abominable wretches'* 
never to be permitted to reside within the haunts of men. 

14. Before leaving the subject of the castelesa tribes I must describe- briefly 
how the names and figures shown in my tabular statement, group I, have been - 
obtained, and in what respect they differ from, those given, in the census report 

(a) Dorru — Under this name no distinction has been made in. the census report 
between Dom the savage and Dom. the musician. The former, is a filthy, degraded 
being, who apes, the Hindu and is regarded with universal loathing. The latter is a 
Muhammadan, holding rather a respectable place in native society, and is by no means 
devoid of culture. It would not be difficult to show how Dom the musician has 
sprang up <jut of Dom the savage. But the two classes are now so entirely distinct 
and have been so for so many centuries, that they must not be confounded. The 
distinctive name of the Muhammadan Dom is Mir&si. 

(6) Bandi.— Only 40 persons have been recorded against this name. Bandi i* 
not a tribe, but an occupation — the occupation of catching live birds for sale. Such 
men generally belong to. the tribe called Banm&nusb, and in this tribe, therefore 
1 have included them. 

(c) Bahrupiya, Bajgi, Nat, Karnatak, Nalak, Bhantu, Sampera. — These seven- 
names do not really represent seven different tribes, as the insignificant number recorded 
against many of them, clearly shows. They may all be considered to belong to the tribe 
generically known as Nat. In iny own table, therefore, I have included them all. 
under this name. They are the jugglers, mimics, actors, acrobats, snake-exhibitors, 
Ac., who wander among the villages of Hindustan and receive fees or gifts in 
exchange for the amusement which they afford. 

Digitized by 


C r > 

(rf) Brijbasi, Gandharp,. KancKan, Tawaif, Nayak, and Neppatar.— These siir 
names may nil be summed up under the common name of Kanchan or Paturiya. In 
my own statement, therefore). I have added up all the figures recorded against these 
names and entered the sum against the name Kanchan. Kanchan^are the here- 
ditary prostitutes of Hindustan— an extraordinary function, whioh. can only, be 
explained by tracing it to the- savage age. 

(e) I have included' in my own list two tribe* which are not mentioned in 
form VIII B. of the census report, tu*., Aheri and Musarha.. The first (Aheri) has 
been avowedly mixed up with Bahelia, and the second (Musarha) with Banmanush 
(see appendix 12: of census report). I think, however, that both of these identifications 
are incorrect. The Aheri is a savage tribe, while the Bahelia is a peaceful caste. 
The Musarha is decidedly less savage than the BanmAnueh and claims to be con- 
sidered as distinct 

15. Castes allied to the hunting state* — Turning- to group HA, " castes-allied to 
the hunting state," we find that the manners and notions of the people included in 
this list come nearest to those of the savage and casteless tribes* named in the pre- 
ceding table. Their status- in< the caste system is so low that it is felmost doubtful 
whether they should be called castes or tribes. My chief reason for putting them in 
the list of castes as distinct from* that of tribes is that they have ceased to hunt the 
jackal, like Kanjars, &c., have abandoned the wandering life, and have made a more 
distinct approach, though still a very distant one, towards Hinduism. They live in 
the outskirts of villages and have long been a recognized element in the constitution 
of the Hindu township, where they are employed as watchmen^ village messengers-, 
Ac. All the castes named in this list are addicted to drunkenness, to the prostitution 
of their daughters, and to- the eating of swineVfTesbj all of which practices are for- 
bidden by Hinduism* Not one of them is allowed to enter a Hindu temple, and no* 
respectable person would touch water drawn by their hands, No Brahman, except; 
those who are set aside as inferiors by their brethren, will enter their houses, or 
discharge for them, even outside of the house, any of those functions for which Brah- 
man* are u&uaHy employed.. All this tends to show that, as compared with, other 
Hindi* castes, they hold, the lowest rank, in the scale of respectability. And this 
tallies with the fact that the stage of culture to- which, they are most nearly allied,, 
namely, the hunting,. is. the lowest and earliest stage in* the development of human., 

16; The names shown in* the appended table (group II A) have been placed as far 
as possible in the order of rank, the lowest and meanest of the castes or tribes being 
placed first, and so on*. The Bauriyas are certainly the lowest class of people named 
in the list. From the local inquiries which I have • made, and from the testimony 
of * writer in the journal of the Asiatic Society of Bengal, who. lived less than a 
century ago, it is clear that their reclamation from a wandering life, in Hindustan at 
least, is a thing; of very recent date. Within the memory of living men they were 
jackal-hunters and wanderers, as the Kanjars, Haburas, and others still are, whom 
I have classed among the casteless tribes, The very names of some of the other castes 
included in group II A disclose the stage of culture, out of which they arose and to 
whioh they are still more closely allied than to any other. Thus Bahelia, the fowler 
easte, is derived from bahali or bahari, a falcon ;. Dhinuk is derived front dhdnu, a 
bow ; and Pasi from pds, a snare. The name Khangar appears to be only a variant of the 
name Ranjar. In character and manners the Khangar is simply an improved Kanjar, 
just as the KhangAr Chattri is an improved form of his ancestor, the Khangar hunter. 
Perhaps the most respectable of the castes or tribes named in this list is the Khatfk,. 
who from a- wandering fowler has developed into a domiciled poulterer,, from a boar- 
hunter into a pig-rearer, and from a hunter and slayer of wild animals generally into 
the caste of butcher to the Hindu community. Of all these castes it must be said 
that,, as hunting and trapping are becoming less and- less- profitable pursuits, owingr 
to the increasing destruction of forest and the constantly increasing pressure of popu- 
lation on the means of subsistence, they are rapidly giving up their original callings* 

Digitized by 


( 8 ) 

*nd taking to agriculture. In almost all parts of Hindustan they are now largely 
employed as field-labourers and sometimes as tenants-at-will on small plots of land. 

17. One thing more must be noted in reference to these castes: it is they 
(together with some of the oasteless tribes which preceded them) who furnish the 
bulk of the criminal classes by whom the peace of the country is disturbed. Hot 
being able, as savage tribes in a purely savage land would be, to make war upon each 
other, they wage a secret war against the community which despises them and to 
which they are still imperfectly assimilated. Under a Government less strong than the 
British, they would at once rise in rebellion against society and revert to their ancestral 
savagery. Let us hear what Colonel Sleeman wrote of the Pasis of Oudh only thirty- 
four years ago, before the Native or Nawabi Government was abolished :— " They are 
employed as village watchmen, but with few exceptions are thieyes and robbers by 
hereditary profession. Many of them adopt poisoning as a trade ; and the numbers who 
did so were rapidly increasing, when Captain Hollings, the Superintendent of the 
Oudh Frontier Police, proceeded against them as thugs under Act III of 1848. * * 
* * These PAsis of Oudh generally form the worst part of the gangs of refractory 
taluqdars in their indiscriminate plunder. They use the bow and arrow expertly, 
and are said to be able to send an arrow through a man at the distance of one 
hundred yards. There is no species of theft or robbery in which they are not ex- 
perienced and skilful, and they increase and prosper in proportion as the disorders 
in the country grow worse. They serve any refractory landowner or enterprising 
gang-robber without wages for the sake of the booty to be acquired. '* ( Vide Tour 
Throvgh Oudh, Vol. I, pages 333—4). What the P&sis were in Oudh only thirty- 
four years ago, the Arakhs, Bhars, Khang&rs, DhAnuks, Ac, were in Upper India 
generally during the anarchy which preceded the establishment of British rule. 

18. I must now explain how the names an.l figures collected in my own table, 
group IIA, have been obtained, and in what respect they differ from those given 
in the census report : — 

(a) B&waria and Boria are entered in the census as if they were different 
castes, whereas they are but one caste with the name differently spelt. I believe 
that the correct spelling is Bauriya. 

(b) Bhar and fiaj-Bhar might, for the sake of simplicity, be amalgamated 
into one, as Baj-Bhar is only a clan of Bhar ; I have therefore included both under 
the name of Bhar in preparing my own statement 

(c) Pasi and Pasia, which are separately mentioned in the census, are really 
one. There is a clan of these called Raj-Pisi, corresponding to the Baj-Bhar, but 
the author of the census report has, rightly as I think, made no separate mention 
of them. 

(d) Tarikash, Eotwar, Pahari, Palm, and Bul&har are not castes, but names* 
of occupations. Tarikash means toddy-drawer ; Eotwar means guardian ; Pahari 
or Pahri (the same name differently spelt) means one who keeps periodical watch ; 
Bol&bar means village messenger. Men employed for such purposes are almost 
always taken from the castes mentioned in this list, and that is my reason for having 
included them in group IIA. 

(e) Ehairwar, Ear&r, and Ehairna, against whom respectively only 56, 436, and 
81 persons have been recorded, cannot, from the smallness of the numbers, be regarded 
as three distinct tribes. They appear to be merely various names of one tribe, which 
is generally known as Ehairwar, and whose habitat is in or near the southern forests 
of Hindustan. I have accordingly included them all under the name of Ehairwar. 

(/) I have not been able to discover any definite clue regarding the Dhingars 
and Dangis, who are said to consist of 1,694 and 3,230 persons respectively. As the 
Dhingar caste or tribe is to be found in Mirzapur, 1 am led to think that they are 
the same as the Dhangars, who are numerous in the neighbouring districts of Bebar 
(see Bengal Census Report, 1872, page 158, where they are classed among the 

Digitized by 


< ■».> 

" aboriginal or semi-aboriginal tribes"). Not knowing what else to do with Dangis, 
1 have pat them among Dbingars or Dhangars. 

(g) No mention has been made in the census report of the caste or tribe 
called Dusadh ; yet the caste certainly exists in the eastern districts of the province. 
In character and habits they are about on the same level with Pasis, Bhars, 
Arakhs, &c. It is said that Dnsadhs formed a large part of the native levies whom 
Clive took into drill, and through whom he laid the foundation of the British Empire 
in India, The military capacities of the Dusadhs are quite in keeping with what 
Colonel Sleeman described of the Pasis of Oudh only thirty-four years ago. 

19. Caste* allied to the fishing state.— This brings us now to the series of castes 
" allied to the fishing state/' all of whom are included in tabular statement, group 
1IB. In point of culture the fishing atateis little, if at all, higher than the hunting, 
and in many countries the two industries have not unfrequently co-existed within the 
same tribe. In India, however, they seem to have kept generally apart, as was the 
case among the native races of North America. The castes of fishermen have ac- 
quired a higher degree of respectability in this country than those of hunters, partly 
because water in the Hindu creed is a much more sacred element than land, partly 
because there is less apparent cruelty in the capturing of fish than in the slaughtering 
of animals, and partly because even the highest and purest castes (including Brah- 
mans) have been compelled to recognize certain classes of men as pure enough to 
draw water for their use • rather than accept the necessity of always drawing it for 
themselves. Thus, considerations of convenience as well as of religion have combined 
to place the Kahar or water-caste on a higher social level than the Pisi or trapper. 
The consequence is that many members of the water-castes have left their original 
calling of fisherman and boatman and have become domestic servants, being called 
Kahars in Hindu houses and Bhistis in Muhammadan and European ones. The trust 
placed in tljem by their masters has refined their manners and given them a taste for 
honesty, which is exemplified in the faot that a man convicted of directly stealing his 
master's money is fined by his fellow-castemen on pain of dismissal from the fraternity ; 
and this is another reason why they have acquired a higher status in society. The 
fishing castes are not debarred the privilege of entering a Hindu temple, as are the 
hunting castes. 

20. In the list of fishing castes, as in that of the bunting, I have mentioned 
the lowest and meanest one first. This is the Meo, whose status and history are 
exceptional. His inveterate propensity to robbery and crime of all kinds recalls the 
savage state from which he has sprung. Disowned by the other castes who practise 
the craft of fishing, disliked and reprobated by Hindus of every class or grade, he is 
turning more and more Muhammadan, and as such he has assumed the name of Mew&ti. 
Further west, in Rajputana and elsewhere, he has taken the name of Mina, which is 
only a variant of Meo, and which, like Meo, signifies " fish. 91 But outside of Hindu- 
stan he is mostly known as a robber, and has almost, if not entirely, abandoned his 
original craft* In Hindustan, too, his propensity to violent crime is well known, 
and during the mutiny he gave us more trouble than almost any other native tribe. 
Like the Plisis of Oudh during the time of the Naw&bi Government, u they formed 
the worst part of the gangs of sepoys in their indiscriminate plunder/' But he is 
still known in Hindustan as a daring boatman and fisherman, and especially as the 
eatcher of river tortoises. I believe myself that the clan of Chattris, called Kachh- 
wfchd or Kachhapagh&ta (tortoise-killer), were originally Meos, and there is na 
doubt whatever that a Meo clan of Chattri* exists at the present day. 

21. As regards the other castes named in this list, there is not much to be added 
to what has been said already in para. 19. The names by which some of them 
are called disclose the stage of culture out of which they sprung and to which they 
are still allied. Thus Kahar means a water-carrier, Mallah means boatman, Kewat 
(a corruption of the Sanskrit Kaivarta) means "one whose occupation is* in water,'* 

Digitized by 


( io ) 

and Dhxmar (anciently spelt as Dhivara) is an old name for fisherman. Bind, Chain, 
and Gaud are tribal, as distinct from functional, names. A special interest attaches 
to the Gaud tribe, as this tribe has, on my theory of caste, formed the nucleus of 
the Gaur Brahmans and the Ganr Chattris, whose name is properly spelt as €t Gaud" 
and not " Gaur." The caste called Dhuriya, Turha or Turi was originally an offshoot 
from the savage Dom, but has now become an entirely distinct class. To each of these 
castes there are sub-castes or clans, whose status in point of culture and respectability 
is not always equal. Thus there are some who keep pigs, like the hunting castes 
named in the previous list, and all such are on the same social level with Pasis, 
Arakhs, Bhars, &c. These would not be able, if they wished, to enter into domestic 
service. Atf the fisherman castes without exception are from their childhood addicted 
to drunkenness — a practice entirely opposed to orthodox Hinduism. 

22. It remains for me to show bow the names and figures given in my own 
table (group IIB) have been obtained, *and to what extent they differ from those 
given in the census report : — 

(a) Gorcha and Gharuk (see names in census, form VIIIB), are subdivisions 
of Chains ; I have therefore included them with Chain. 

(b) Eadhars and Tatwas are nearly, if not quite, identical with Mallahs; I have 
therefore made no separate mention of them in my list, but included them amongst 

(c) Turha is the same caste as Dhuriya or Turi and has been included with it. 

(d) Singhar is not a caste, but a name which could be applied to any one who 
cultivates the singh&r or water-nut, as all the castes named in this list are in the 
habit of doing. 

(*) Kahar. This is an ambiguous word. Generically, it is applied to all the 
castes named in this list (except the Meo), because individual members of these castes 
can and sometimes dp enter into domestic service as Kahar or " water-carrier." 
Specifically, the name is given to that caste which has entirely and for many gene- 
rations continuously withdrawn from the occupation of fishing, boating, Ac., and 
which now exclusively devotes itself to domestic service. Kahar in the former sense 
is not the name of any one caste, and should therefore not appear as such in the list 
Kahar in the latter sense comes under a totally different category and belongs to the 
list of serving castes for whom a separate tabular statement (group V) has been 
provided ; I have therefore omitted the name Kahar altogether in/ the statement 
ealled group IIB. 

(/) There aro four well-known castes, of whom no separate mention has been 
made in the census report : these are Eewat, Gaud, Gaudia, and Dhimar. The author 
has pointed out in a note to page 136 of his Preliminary Dissertation that Gaud, which 
he incorrectly spells as Gond, has been included among Kah&rs, because the " Gond" 
of the Gangetic valley must not be confounded with the " aboriginal Gond" of Cen- 
tral India, as was done in the previous census. The names, however, are not the 
same. One is Gaud and is pronounced in English as Gaur. The othdr is Gond and 
is pronounoed as Gond. The tribes are certainly very distinct, but as long as each 
appears in its proper place and with its appropriate spelling, Gaud in group IIB 
and Gond in group I., no confusion can arise. 

23. Castes allied to the pastoral state.— Ascending a few steps higher in the social 
ladder, we come to those castes which are " allied to the pastoral state." The occu- 
pation of cattle-grazing, as I need scarcely add, represents a higher grade of culture 
than that of bunting or fishing : and hence the pastoral castes stand on a higher so- 
cial level in the Hindu community than the castes which preceded them. The change 
from the hunter to the nomad, and from the wandering cattle-grazer to the stationary 
one mast of course have been gradual : and the first nomads were almost as much addict- 
ed to robbery and bloodshed as men who lived by hunting and fighting only. This 
is true at the present day, for example, of the Bedouin tribes of Arabia and Egypt, 

Digitized by 


( 11 ) 

and some of the Tartar tribes of Central Asia. Every phase of nomadism is repre- 
sented at the present day in India, as I shall presently show, by some corresponding 
caste ; and in the tabular statement annexed (group IIC) I have given the names of 
these castes, as far as possible, in the order of that phase of culture to which they are 
most nearly allied, the lower being mentioned first and the higher last. 

24. If the reader will refer to the tabular list (group IIC), in which these castes 
have been collected, he will observe that the names by which they are called, indicate 
very clearly the stage of culture, out of which they originally rose and to which as 
castes they still approximately correspond. Thus Goshi is derived from Gosha, a 
cattle-pen. Gadaria, the shepherd caste, is derived from Gadar, an old Hindi word for 
sheep. Ahar and Ahir are variants of the Sanskrit Abhir, which means nomad, "one 
who wanders about." This caste or tribe was known to Pliny under the name of 
Abhiri. Gujar is simply a variant of Gochar, which means "cattle-grazer." The 
Jit easte has retained its tribal, instead of taking a functional, name, on account of 
the fame of the great popular hero, Krishna, who was born in its ranks. 

25. As was stated above, there are various gradations of rank, corresponding to 
the various grades of nomadism, among the pastoral castes of India. Thus the Gaddi, 
or Goshi, who has been placed first, that is, lowest, in my tabular list, is not merely 
a cattle-grader but a cattle-lifter, and he recalls in other respects the predatory or 
earliest phase of nomadism. If he can find space, as he still does, on the sides of the, 
Vindhya hills, he prefers to this day the purely migratory to the settled life. The, 
Hindu caste is generally called Gaddi, and Ghosi is the Muhammadan counterpart 
Amongst Muhammadans the Ghosi ranks quite as low as the Gaddi amongst Hindus, 

' and for the same reason. The very name " Gaddi" appears to be merely a variant of 

Gadiya, Gidiya, or Gandila— a tribe of savages and cutthroats, of whom only a few 

are now to be found in Hindustan, though they are said to be still numerous in the, 

Fanj&b as a clan of the predatory Bansis. Thus Gaddi the cowherd is merely an. 

improved form of Gadia the savage, just as Khang&r is an improved form of Kanjar, 

Dhuriya of Dom, Ahir of Banm&nush, Ac. A man. of the Gaddi caste cannot enter a . 

Hindu temple, which proves that his status in society is as low or almost as low as that, 

of the hunting tribes or castes. The castes standing next to the Ghosi or Gaddi in, 

order of respectability are the Ahar and the Gujar. The former is by name merely. 

a variant of Ahir, but in character he is less civilized. He therefore constitutes a 

distinct caste and is recognized as Wh by all Hindus. The Gujars, especially those 

in the Meerut Division, are notorious for their predatory habits. In the Moghal times 

they were largely employed as watchmen, a clear proof of their thievish and lawless 

propensities ; and in the time of the sepoy mutiny they gave as much trouble as the 

Meos, plundering friend and enemy indiscriminately. The Ahir is very different. 

He is as a rule orderly and peaceable wherever he lives ; and from a wandering 

cattle-grazer or nomad, which (as his name implies) he originally was, he has for 

several generations past been settling down as a stationary husbandman, and this 

not so much from choice as from the constantly increasing scarcity of fields or , 

forests suited to his ancestral calling. The Gadaria resembles the Ahir in character 

and pursuits, except that he grazes sheep and goats instead of cows and buffaloes. 

He is largely occupied in wool-shearing and wool-weaving, and he adheres to this 

calling as long as he can in preference to agriculture. The Rew&ri or camel-breeder 

is a respectable caste, belonging to Raj pu tana rather than to Hindustan. The Jit 

stands higher in the social scale than any of the other castes belonging to this group. 

The name J&t or J&d is simply the modern Hindi pronunciation of the Sanskrit 

Tadu or Jddu, the tribe in which the great popular hero, Krishna, was born. In the 

Frem S6gar and elsewhere the Tadu tribe is described as simply pastoral, and the 

flirtations of Krishna with the milkmaids would lose all force and significance on any 

other hypothesis. Krishna, then, was a great nomad king, who achieved renown aa 

a victorious warrior no less than as an amorous swain ; and the tribe in which he was 

born became, owing to the fame of his greatness, the highest and most powerful 

Digitized by 


( 12 ) 

of the pastoral tribes of Upper India, as it has remained to this day. His modern 
worshippers and admirers, not satisfied with thinking of him as a eowherd, have raised 
him to a Raj pot; and hence in the western districts of the Panj&b no clear distinc- 
tion between J&ts ancf R&jputs can now be made out. In Hindustan, however, the 
Jats are still many degrees below Chattris or Rfijputs. Nor have they entirely thrown 
off the nomad and predatory instinct : for, as Mr. Beames remarks, " not long ago they 
were notorious cattle-lifters, nearly as bad as their neighbours, the Gajars" (Elliot's 
Supplemental Glossary, Vol. I, page 136). As a caste they still retain many of the 
customs and traditions peculiar to the pastoral state ; but their present occupation is 
chiefly agricultural, and in the Meerut Division they are said to raise finer crops of 
sugar than any of the agricultural castes proper. 

26. A brief allusion must be made in passing to the theory repeated by 
almost all writers on Indian ethnology > that the pastoral castes, especially the 
J&t, the Gujar, and theAhir, are of Scythian origin, that is (I presume) of neither 
Aryan nor indigenous blood. Several reasons have been alleged in support of 
this view. Speaking of the J&ts, Mr. Williams remarks that " the custom of 
the kar&o, the marriage of a widow with the brother of a deceased husband, 
clearly indicates a Scythian origin'* (Oudh Census Report, 1869, Vol. I, page 
93). Again, the same writer speaking of the Ahirs (see page 100) gives some coun- 
tenance to the notion that the word Ahir is derived from ahi> a snake, and considers 
that the regard for snakes shown by this caste " must be a relic of the old serpent- 
worship, and seems to confirm the supposition of a Scythian origin for this caste." 
In using the argument drawn from the remarriage of widows Mr. Williams was only 
following Sir Henry Elliot, who remarks that " Ahirs conform to the customs of 
Gfijars and J&ts in respect to the marriage of the elder brother's widow" (see Sup- ' 
plemental Glossary, Vol. I, page 5) ; and in several of the Gasetteers of the North- 
Western Provinces lately published allusion is made to the community of this custom 
as a mark of community of origin. The ethnology of th^ Giijars is learnedly dis- 
cussed by General Cunningham, who identifies them with the Tochari, alias Yuchi, 
alias Kushan, alias Kaspiroi, alias Tbagarii, alias Kuei-shwang, alias Korson, Eorsea, 
Khuransu, or Koruno ; and in this way they are lodged at last in the city of Khu- 
rasan and traced to a Scythian origin (see Archaeological Report, Part IV, pages 
25-29). Independently of this, Sir Henry Elliot had already identified the J&ts with 
the Getoe or^Massa-Getoe of ancient Scythia, and these again with the Yuechi, Yuchi 
or Yuti, whom General Cunningham has identified with the Gujars (Supplemental 
Glossary, Vol. I, page 133). Finally, the Scythian origin of the J&ts has been accepted 
as an historical fact by Dr. W. W. Hunter > who includes the Niga tribes also in 
the same category; for Ndga, as he shows, means snake, and " serpent-worship formed 
a typical characteristic of the Indo-Scythic races " (Indian Empire, Chapter VII, 
page 173> edit. 1882). In his " Brief History of the Indian People," which is the text- 
book for the entrance examination to the Calcutta University, several thousands of 
youths throughout the presidency of Bengal are now being taught that the Indian 
population is made up of three distinct racial elements— the Aboriginal, the Aryan, and 
the Scythian. In Chapter VIII of this volume he remarks :— " The third race, the 
Scythian, held a position between the other two. The Scythian hordes who poured 
into India from 126 B. C. to 400 A. D. were neither hunters like the Indian non-Aryan 
tribes, nor cultivators like the Aryans. They were shepherds or herdsmen who 
roamed across the plains of Central Asia with their cattle and whose one talent 
w as for war." 

The grounds, then, on which the Scythian origin of the pastoral oastes has 

been maintained are three in number : (1) the custom of the kardo ; (2) the worship 

u Should a younger brother haTe be- of snakes ; (3) the identification of proper names. 

eHeTb^ The first argument is worth nothing, because the 

must then be made equally between that marriage of the elder brother's widow was a 

son, who represents the deceased, and 

his natural father : thus is the law set- Hindu custom legalised by Manu's Code (see 

extract in margin), and though it has since been 

tied."— Chapter IX, section 120. 

Digitized by 


( 13 ) 

prohibited to the -higher castes by the code of Parfisara, it is still practised by all the 
lower castes, and is not, nor ever has been, confined to the Q ajar 8, Ahirs, and Jits. 
The second argument is likewise worth nothing, because snakes are worshipped all over 
India by all castes alike, and have at different times been worshipped all over the world, 
wherever snakes are to be fonnd. The third argument, (the identification of proper 
names) is ingenious,- bat not convincing. The reader is already aware that Jit is sim- 
ply the modern Hindi pronunciation of Y&du or Jadu, that Ahir can be easily traced 
to the Sanskrit Abhir, and that Gajar is merely a variant of Gochar or 'cattle-grazer. 

Even if it could be proved that these castes were of Scythian blood, the result 
would only strand us in deeper ignorance than before; for the best authorities are 
not agreed as ito who the Scythians were, some contending that they were Aryans 
and others Mongols. Considering that these so-called Scythian tribes of India are 
in physical characteristics precisely similar to the rest of the Indian population, it 
is vain to expect one to believe that they are of an alien ethnical stock. Moreover, the 
the pastoral castes are the necessary intermediate link between the hunting and the 
agricultural ; and this fact alone, unless we are to discredit the analogies of history 
and the conclusions of science, is sufficient to prove that they are not of foreign but 
of indigenous blood, 

27. The differences between the names given in the census report and those 
shown in my own table (group I IB) are in this instance very slight :— 

(a) Gaddi has been entered as an equivalent name to Goshi, as the former is the 
name of the Hindu caste and the latter that of the Muhammadan counterpart Even 
Ahirs and Gfijars who have become Muhammadan frequently call themselves by the 
name of Goshi, 

(b) Baona, against which only X43 persons have been recorded in the census, is (so 
far as I can learn) a sub-class of Gaddi, amongst whom, therefore, I have included them. 

(e) Dhandhor, against which only eight persons are recorded, signifies etymologi- 
cally "one whose wealth lies in cattle," being derived from rfAan, wealth, and dhor, 
cattle. It is not the name of a caste. Such a name might be applied to any one 
who lives by cattle. But I learn that it is not unfrequently used as an euphemism.for 
cattle-lifter ; I have therefore included them amongst Gaddis. 

88. Agricultural ca$tes.^-We come now to the agricultural castes proper — 
those whose sole or chief industry from several centuries past has been the cultivation 
of the soil. Agriculture represents, as is well known, a higher phase of industry 
than either hunting or nomadism, nor is it difficult to see how this industry arose out of 
the two which preceded it. So long as man subsists by hunting game, catching fish, 
gathering wild fruits or seeds, and digging bulbs and roots out of the earth, he feeds 
himself much in the same way as the birds and beasts. But natural intelligence must 
soon suggest to him that if seeds or roots are put into the earth by his own hands, 
they will grow there quite as readily as if they were sown or planted by nature. 
The same instrument which he uses as a spear for hunting game serves equally well 
for dibbling seeds into the soil. Again, to plant a root is much the same kind of pro- 
cess as to dig one up ; and here, too, the same instrument which he uses as a pick or 
hatchet serves also as a hoe. Such is the form of agriculture coeval with the hunt- 
ing and fishing states. A higher stage begins to be reached when men by invert- 
ing the hoe and giving it a handle change it into a plough. Having learnt to tame 
and rear cattle for use, instead of hunting them for game, it is not long before they 
find out how to attach them to the plough, instead of dragging the plough through 
the earth by their own labour. Agriculture at this secondary stage is chiefly noma- 
die; that is to say, men seldom cultivate the same piece of land for a second year ; 
they are owners of the crop raised on it, but not of the soil itself. It is only when 
men remain permanently on the same patch of land and have acquired the right of 
permanent occupation and of transmission to their children, that the agricultural 
state proper has been reached. It is then that villages and towns are formed ; new 


Digitized by 


( U ) 

arts and industries spring up ; trade finds for itself new channels ; and new ideas of 
duty public as well as domestic, are suggested to the mind. The agricultural state, 
then! ranks decidedly higher in the scale of human progress than either the huntmg 
or the merely pastoral states. By the same law the agricultural castes in Tndia 
ipso facto represent the agricultural state, stand higher in the Hindu social scale 
than the castes preceding them. 

29. The names by which most of the agricultural castes are called (vide group 
1ID) tell their own tale. Lodha signifies " clod-breaker," being derived from lod, 
a clod of earth, and hd, breaker. Kaodu is derived from kdnd, a river-bank. Simi- 
larly Kachhi is derived from kaehh, the alluvial soil on the banks of rivers, which, on 
account of its lightness and the superior facilities for. irrigation, must have come un- 
der the plough at an earlier date than the dryer and harder soil of the uplands which 
were once covered with forest. Kurmi is derived from Kurma, the tortoise incarna- 
' tion of Vishnu, which is specially worshipped by agricultural castes, as the fish that 
supports the earth. Tamboli is derived from tambol, the pan creeper, and the name 
is well suited to the caste specially employed in rearing this delicate herb. Mali is 
derived from maid, a wreath of flowers-an appropriate name for the caste which has 
made floweY-gardening its peculiar function. Bhuinhar is derived from bhu or bhdmx 
(the earth) and hdr (taker or possessor), and is very properly applied to a caste the 
members of which pride themselves upon being owners or landlords, and not merely 
tenants of the soil they cultivate. Thtis, if other evidence were wanting, the names 
by which the agricultural castes are called show clearly enough to what stage of 
industry they belong. The generic name of most of these castes is Kisban or plougher, 
as Gwala or milkman is the generic name of the pastoral castes, and Kahar or water- 
man of the fishing castes, and Ohirimar or fowler of the hunting castes. 

30. As in the case of the pastoral castes .already described, so among the 
agricultural castes proper, there is a graduated scale of dignity whieh is recognised 
by the general Hindu community, and which we must now endeavour to account for. 
Omitting the Bhot and the Kamboh-both of foreign origin and neither of much 
importance -the castes enumerated under this heading stand as follows in the order 
of dignity the lowest being mentioned first and the highest last :— 

' i Bayir. 

1V * •" I Bhuinhfcr. 

i K&ndu. 

II. ... iK4chhi. 


31. The Lodha* and Bayar not only stand lowest in the list of agricultural 
castes but they are socially inferior to the best of the pastoral, just as the Gaddi and 
Goshi] the lowest of the pastoral castes, rank below the best of the fishing or water- 
man classes. Nor is this at all inconsistent with our general theory of gradation. 
For the course of human progress is spiral rather than rectilinear, because humanity 
advances only by a series of reactions against the point of vantage last gained. 
The traditions and sympathies of the Lodha connect him very clearly with the semi- 
savage Musarha, whom I have placed among the "castehse tribes"— from which it 
might be inferred that his adoption of the agricultural status is comparatively recent 
as compared with that of others. And this inference coincides with the fact ; for 
it is well known to those who have lived in Oudh that there are some families, of 
Lodhs or Lodhas, who to this day prefer the rougher task of felling trees and making 
the first clearance of land to the tamer and more monotonous industry of tilling 
the open plain. The Bayar holds about the same rank as the Lodhd. His name 
is identical with that of an old tribe of freebooters, who are celebrated in local tradi- 
tions as builders of mud forts and founders of petty kingdoms. Both of these castes 
are still occasionally addicted to turbulence— a mark of their more recent reclama- 
tion from savagery. 

32. The three next castes— the Xandu, the Kurmi, and the Kachhi— are peaceable 
and industrious, wherever they are to be found. Their status is about on a par with 

Digitized by 


( 15 ) 

that of the best of the pastoral castes (the Jat and the Ahir), hot decidedly above 
that of the rest. The Kandu is an offshoot from oertain fishing tribes, and like 
them cultivates the singhdra or water-nnt. He also raises crops of grain, especially 
of rice ; and from a rice*grower he has developed in some places into a rice-parcher, 
for the sale of which he sometimes keeps a shop. The Kurmi (called also Kunbi) 
is the greatest grain-growing caste in India, and he has been deservedly called 
" the backbone of the country." The women and children take almost as active 
a part in the labour of the fields as the men. The British Government' has 
generally recognized his superior skill as a cultivator by fixing a higher assess- 
ment on his lands than would be demanded from a Brahman or Chattri occupant. 
The Kachhi sometimes grows grain like the Kurmi, but is chiefly known as a market- 
gardener. Some of the most prosperous members of this caste hold large plots of 
land in the vicinity .of towns and cities, where they raise crops of poppy, potato, 
tobacco, sugar, &c. The name " Kachhi," as we have shown, is functional and not 
tribal. But another name, Koeri, by which he is sometimes called, shows that the 
wandering and semi-savage Kol, who is still to be found in the districts south of the 
Ganges, was his remote ancestor, though every tradition of such descent appears 
now to have been forgotten. Another synonym, Murao, appears to point to some 
other ancient, but now forgotten, tribe as the joint ancestor of this useful and indus- 
trious caste. 

33. The five castes whom we have named thus far still indulge in the flesh of 
field-rats, a habit which tends to lower them in the eyes of their fellow-countrymen, 
and connects them remotely with the Ugras, Pukkasas, and other savages who are 
denounced in Manu's Code as " hunters of animals that dwell in holes." But the 
four castes remaining to be described abstain entirely from such unclean diet, and this 
is one of the reasons why they hold a higher rank than the preceding. The Tam- 
boli and the Mali stand next above the Kurmi and Kachhi. The first is the well- 
known and respected caste of pan-grower. In the rearing of this plant they enjoy 
an exclusive monopoly, which must have taken its rise from the fact that only the 
son of a pan-grower him9elf can acquire the secret of rearing such a delicate herb. 
It is the custom of this caste not merely to grow the plant, but to prepare it for sale, 
and to keep a shop for this purpose. If the trade prospers, the Tamboli takes to 
money-lending and usury, the trade proper of the Oswals, Agarwals, Khatris and 
others : and I feel no doubt that these wealthy and distinguished castes have at 
various times been recruited from pan-growing families. The Mali is the well-known 
caste of flower-gardener. His art, like that of the Tamboli, indicates a degree of 
refinement above that of the mere grain-grower and vegetable-gardener. But a re- 
fined art leads to a corresponding refinement in manners, feelings, and mode of life 
generally, and any caste which practises such art or arts rises to a proportionate 
height in the social scale. Such has been the case with the Mali, and such is the 
theory of caste — gradation assumed in this paper. Another fact whioh has raised 
the Mali above his fellows is one connected with religion. To Hindus of all ranks, 
including even the Brahman, he acts as a priest of Mahadev in places where no 
Goshayen is to be found, and lays the flower-offering on the lingam by which the 
deity is symbolized. As the Mali is believed to have some influence with the god to 
whose temple he is attached, no one objects to his appropriating the fee which is 
nominally presented to the god himself. In the worship of those village godlings 
whom the Brahman disdains to recognize, and whom the Goshayen is not permitted 
to honor, the Mali is sometimes employed to present the offering, fie is thus the 
recognized hereditary priest of the lower and more ignorant classes of the population. 
It is not at all improbable that certain Mali families have at times gained admission 
into the ranks of Brahmanhood. 

34. This brings us lastly to the two* highest of the agricultural castes, ' the 
Tagd and the Bhuinhar. There are two great facts which place them above all the 
preceding castes. Firstly, they do not allow the remarriage of widows, whereas everjr 

Digitized by 


( 16 ) 

caste hitherto named in this paper permits the practice. This shows that they have 
been for a much longer period under Brahmanical influence, and they have acquired 
an increased dignity in consequence. Secondly, they are generally owners, and not 
merely tenants, like the castes previously named, of the land on which they raise their 
crops. In this respect they make a much nearer approach to the status of the Chattri 
or landlord caste, which ranks immediately above them. Almost all cultivators except 
the Tag& and Bhuinhar found it necessary to place themselves under the protection 
of some Ohattri landlord, who, being of the warrior caste, protected them from aggres- 
sion from without, on condition of rsceiving rent in return : whereas the Tag& and 
Bhuinhar have been their own landlords and protectors from the beginning. The. 
one fact which places these two castes in a rank below that of Chattri and Br&hman 
landowners is that they till the soil with their own hands, whereas the Chattri and 
Brahman are compelled to employ hired labour for the purpose ; the former thinking 
it beneath his dignity, and the latter iagainst his religion, to handle the plough. The 
Ta<r& hunts the wild-boar and drinks spirituous liquors, just as the Chattri does. The 
Bhuinhar leans rather to the Brahman than to the Chattri type. In places where the 
Brahman element predominates he abstains from boar-hunting and wine, wears the 
Brahmanical thread, attends Brahmanical festivals, and adopts the same marriage 
ceremonies as Brahmans. To this day no Hindu can tell you distinctly whether a 
Bhuinhar is a Brahman or not: and there is no reason to doubt that many of the 
modern Brahmans are sprung from certain clans of Bhuinhars, who at some earlier 
age renounced affinity with their own brotherhood and entered into the ranks of the 

The gradually ascending scale of rank which we have now traced among the 
several agricultural castes, beginning with the Lodh who was lately a savage of the 
woods, and rising by degrees to the Bhuinhar, who is almost a Brahman, illustrates 
on a smaller scale the process by which the entire system of castes throughout the 
Indian continent was brought into existence, and exemplifies the principle by which 
their several ranks in the social scale have been determined. 

35. It remains to state briefly in what respects the names of the agricultural 
castes shown in my own table (group IID) differ from the names given in the census 

(a) Kachhar and Kachhwa, against which only 290 and 1,587 persons have been 
recorded respectively, are not names of separates castes, but merely variants of the 
name Kachhi ; 'I have therefore included them in the same total. 

(6) Garaela, Ehagi and Setwar, against which 1,754, 38,007, and 26,498 persons 
have been recorded respectively, are names not of separate Castes, but of sub-castes 
of Kurmis ; I have therefore included them all under the name of Kurmi. 

(e) Bot, against which only 3,191 persons have been recorded, is not a separate 
caste, but merely another way of spelling or pronouncing Bhotiya, against which 
9,205 persons have been entered. Originally immigrants from the northern hills, they 
have for several centuries past been domiciled in the plains, but still retain the features 
of hill men. 

(d) Mai, against which only 3,218 persons have been recorded, is the same as 
Mali or Gardener. 

(e) Pattiar is the only agricultural caste named in the census report which I 
have not been able to identify. Perhaps it was written for pattidar, or co-sbarer of a 

. landed property. If this was the case, the 547 persons recorded against the name 
(all of whom, too, are said to belong to a single district, Fyzabad) cannot be called 
a caste. 

3o\ The landlord and warrior caste. — We come now to the fifth and last of the 
series of "castes connected with the land." This is the landlord or warrior caste 
whose name is variously known as BSjput, Chattri, or Thaknr. The function of this 

Digitized by 


(17 ) 

caste, if we look to its original status, is to protect and rule all other castes, receiving 
rent and taxes in return. The Chattri proper is the landlord, warrior, and ruler. 
Such a caste cannot but take precedence, in point of dignity, of all castes and classes 
of men subject to its authority ; and hence the M&nava Code has assigned to it a rank 
second only to that held by priests or Brahmans, who by the same code are the advisers, 
secular as well as spiritual, of the kings and rulers of the earth. 

37. In every part of the world, wherever agricultural settlements have .super- 
seded nomad hordes, the landlords or landed aristocracy have led and organized the 
armies of the State and have constituted pre-eminently the military class or caste. 
India has been no exception to the rule. The very name Chattri (or Eshatriya as it 
was originally spelt) implies this. For though the name " Kshatriya " has been 
universally used to designate a warrior, its primary meaning is " landlord," being 
derived from Kshatra, which signifies the possession of a territorial domain. One of 
the other titles, " Rajput, " signifies " a man of royal blood '* — a further indication of 
the original identity of the landowning and militant functions. This double mean* 
ing of warrior and landlord within, a single word finds its analogue in the language 
of the Greeks, amongst whom the landed aristocracy were called the aristoi, the 
root.of which is identical with that of Ares, the god of war. The term " baron," 
which in most countries of Europe was given to the owners of large landed estates, 
is derif ed from a root signifying manliness and valour; and it is needless to add that 
these territorial magnates were distinguished throughout the twiddle ages, like 
the Indian RAjput, for their love ef fighting and high sense of honour. The king 
himself was only a baron on a larger scale, just as the Indian r&ja was merely a 
magnified Chattri. 

38. The fact of the names of this caste being purely functional, and not tribal, 
gives no countenance to the theory so commonly expressed by writers of established 
reputation, that Chattris are the direct descendants of the Aryan tribes who oame 
from Central Asia, while the castes below them afe of aboriginal or mixed blood* 
And if space permitted, it would not be difficult to show that this theory is opposed 
both to the testimony of the Hindus themselves and to the teachings of history. " In 
the savage state," says Mr. Herbert Spencer, "there are no owners of the tract 
occupied, save the warriors who use it in common for hunting. During the pas- 
toral life good regions for cattle-feeding are held jointly against intruders by force 
of arms. And where the agricultural state has been reached, communal possession, 
family possession, and individual possession have from time to time to be defended 
by the sword. Hence in the early stages the bearing of arms and the holding of 
land habitually go together." Now when we reflect that each of the three stages 
alluded to in this passage have existed in India for some thousands of years, and are 
represented even to this day (as. I believe I have proved in this paper) by corres- 
ponding castes, it might be assumed on a priori grounds that each stage of industry 
has contributed its quota of victorious chieftains, and that the caste of Chattris is 
simply a congeries of men, of any tribe whatever, who were able at various times to 
seize lands and keep them, and who by intermarriages and alliances with others of 
their own status built up by degrees a separate class or caste distinct from and above 
the rest of the community. This is exactly what we find to have been the case in 
reading the histories and names of many of the Chattri families given in the pub- 
lished Gazetteers .of these provinces. Among the names of these families or clans 
we find BarwAr, Kbangar, Gaharwdr (a variant of Khairwar), Gadariya, Gujar or 
Bhat-Gujar, Yaduor Yadon, Meo, Oaud(Anglice Gaur), Chamar-Gaur, Jaiswdr, Bandel, 
Domwar, Kh&gi, Nig or N&gbansi, Baheliya or Bheriya, Gadiya, Ac. The reader, who 
has attended to the names of the hunting and pastoral castes already described, will 
perceive at once the identity between these names and those of the Chattri clans just 
quoted ; and such identity shows clearly enough what the origin of many of the modern 
Chattris really was. Any clan, which had gained possession of estates and was able to 


Digitized by 

oy Google 

( is y 

bequeath them to their posterity, came as a matter of course under the patronage of 
Brahmans, and its leading men were thus gradually educated into kinship with the older 
members of the aristocratic caste, who had received a similar education before them; 
A caste which lives only for territorial dominion and military renown cannot afford to 
be exclusive. It thrives only by expansion — that is, by incorporating within its ranks 
men, of whatever tribe, who have proved their fitness by force of arms and whose 
alliance would add to the power and wealth of the fraternity. If men of the modern 
Chattri caste were the pure-blooded descendants of the ancient Arya, it is a singular 
circumstance that so mauy of these Aryan knights, who are supposed to have come 
from some country west of the Indus, should have assumed the tribal names of the 
indigenous Indian savage, and that out of the 99 clan names collected by Mr. Sher- 
ring there should be only one (Bhrigu-bansi) which can be traced to the Vedio age. 
Once educated and civilized after the Brahmanical model, the territorial chieftain, 
whatever his ancestry may haye been, becomes a Chattri nobleman, and with the 
help of some Bhat or bard or other mendacious parasite acquires a genealogy, which 
raises him to a position in his own country similar to that held by the Spartiatae in 
Lacedsemon, the Aristoi in Athens, the patrioians in Rome, or the barons in Europe. 
In Upper India the manufacturing of Chattris is a process still going on before our 
eyes, and what is happening now has been in operation for the last two thousand 
years at least For example, the Janw&rs of Oudh held till lately a very obscure 
status as Rajputs, and on account of the doubtfulness of their origin they were com- 
pelled to be content with local alliances. But now they have begun to marry their 
daughters to the Rathors andKachhw&h&s of Raj pu tana (Oudh Gazetteer, Vol. II., 219). 
A writer in the North- Western Provinces Gazetteer (Vol. VIII., page 73), speaking of 
the Rajputs of the Muttr'a district, remarks that " about six- sevenths of them are of 
impure blood, and are not admitted by the higher clans to an equality with them- 
selves. The crucial test," he adds, " of purity of blood is th9 rejection of the custom 
of karao (remarriage of widows)." But such a test, far from proving or disproving 
purity of blood, merely shows whether the clan has or has not become perfectly Hin- 
duized according to modern notions ; and three or four generations are quite enough 
to turn an Indian savage into an orthodox Hindu, if the opportunity comes in his 
way. Moreover, I have proved already in para. 26 that in the time of Mann's Code 
the remarriage of widows was under certain conditions prescribed as a duty by 
Brahmanical law. 

There is strong reason to believe that most of the Oudh Chattris are of Bhar 
blood, but it would be too lengthy to quote the evidence in this place. " There are 
two Bhar landlords/' writes Mr. Sherring, " in the Mirzapur district. But these 
men, disloyal to their tribe, though wise in their generation, feeling the grievous 
burden of their* social position, affect a Rajput title, notwithstanding that it is well 
known that they are descended directly from the Bhars " There is nothing to pre- 
vent suoh men, two or three generations hence, from buying husbands for their 
daughters from some scion of the royal tribes, if they have sufficient wealth. It is 
a mark of honor in all castes, and especially among Chattris, that a man must seek 
to marry his daughter into some clan higher than his own : and this custom, which 
has been in force for the last two thousand or three thousand years, has heen one of the 
causes at work by which families of the lowest origin have been slowly but surely raised 
into the ranks of the landed aristocracy, until at last all distinctions of race and blood 
have become obliterated. The Chattri himself is merely a Tag& who disdains to hold 
the plough. The Taga is merely a more cultivated form of Eurmi. The Kurmi is 
only an improved form of Lodh. The Lodh is not many degrees above a savage 

39. In speaking of the Chattri or Rajput as the ruling and warrior caste, 
1 was of course referring to its original function rather than to its present status. 
Between the ideal Chattri of the old Hind a period and the Chattri of the present day 
a wide difference exists. Its function as the ruling caste received a shock from the 

Digitized by 


( » ) 

Muhammad an conquest, from which it never recovered : and many Chattris of high 
tank have at various times embraced the creed of Islam in order to retain their 
estates. What survived of this function was annihilated by the establishment of 
British rule. The other function, that of fighting, still to a certain extent remains ; 
for the native regiments, through whom the Mahrattas were defeated and the provin- 
ces of Upper India added to the empire, were largely recruited from the ranks of 
the warrior caste. But the Pax Britannica is rendering even this function almost 
a thing of the past The Chattris in fact have seen their best days, and for several 
centuries past have been, a down-going race. They have not had sufficient intellec- 
tual .keenness to compete with Brahmans and Eayasths under the new order of things, . 
and a large number of their estates have passed into the hands of successful plead- 
ers and merchants. Many of those who have managed to keep their lands will be- 
fore long be compelled to hold the plough for themselves. 

40. In the census report two entries are made on account of this caste — Chau- 
han and Rajput. Chauhan, however, is only one of the numerous clans of Chattris 
or Rajputs, and if one clan is to be separately mentioned, the same should be done 
with all. I think, however, that in a census of castes, where simplicity is so much 
to be aimed at, the generic caste name is sufficient. Detailed information respecting 
the clans is given in the Gazetteers of the different districts. 

41. This completes the long series of "castes connected with the land," of 
which a brief description has been given under the five subordinate headings of hunt- 
ing, fishing, cattle-grazing, agriculturist, and landlord. It is not surprising to find 
that this large group contains 19,064,236 persons, or nearly one half of the entire 
population of the province. As we have already more than once implied, the original 
distinctions between hunter, fisherman, cattle-grazer, and agriculturist are becoming 
more and more effaced as time goes on ; and all industries connected with the land are 
becoming more and more merged into that of agriculture. Such names, then, as hunter, 
fisher, &c., denote the origin and explain the social status of the castes corresponding 
rather than describe the functions or industries which they chiefly practise at the present 
day. It is curious to note, however, how the original functions peculiar to these several 
castes have survived in tradition. Thus, for example, at the present day, certain hunting 
and pastoral castes, such as the Musarha and the Gaddi, are in the habit of parcelling 
out wide tracts of country among themselves as if they were still owners of the soil, for 
purposes of root-digging, hunting, fruit-gathering and cattle-graztng ; and when an 
appropriation of this kind is given as a daughter's dowry, the monopoly so conferred 
is rfespected by the rest of the fraternity. Again, there is no such thing in India as a 
farm, in which glebe and pasture lands are Maintained side by side as parts of a single 
holding ; and the same man, whatever his caste may be, never oobbines ploughing 
and cattle-rearing as parts of a common industry. It appears, then, that the tradition 
of the original diversity between the pastoral and the agricultural states and between 
the castes corresponding has been strong enough to keep the two industries apart up 
to the present day ; whereas in every other country, where agriculture has reached 
the same stage of advance that it has done in India, they have become completely 
amalgamated, the one being considered incomplete without the other. 

42. The artisan castes. — We now come to a series of castes whose functions are 
altogether distinct from those of the preceding. These are the artisan castes. The 
generic name by which they are known in Upper India is that of Karigar. Here an 
entirely new law comes into play. It is now a thoroughly established fact that no 
nation or people can make any material advance in civilisation until they have dis- 
covered, or at least learnt to imitate, the art of metallurgy, and the ignorance of this 
art is the invariable mark of backward races all over the world. The hunting, fishing, 
and cattle-grazing industries can be and are carried to a high degree of efficiency 
without the use of metals. But agriculture cannot be practised, in any but the lowest 
and most rudimentary form, without the help of iron tools and implements ; and until 

Digitized by 


( 20 ) 

such appliances have been provided, hunting, fishing, or cattle-grazing will continue 
to be the predominant industry. Hence, roughly speaking, we may lay down the la* 
that the agricultural state proper co-exists with, and cannot precede, the age of 
metallurgy, while the hunting, fishing and nomad states precede the age of metallurgy 
and are coeval with the inferior stages of art. 

43. The same law has determined the rank or social status of the several artisan 
castes of India. It will be invariably found, if we look into the status of each of them 
individually, that those castes or trade-unions, which reptesent the inferior stages of 
art, stand lower in the social scale— that is, receive a less degree of respect from the 
Hindu community generally — than those which represent or are coeval with the metal- 
lurgy stage ; and that all such castes are about equal in rank, as members of the 
Hindu community, to those castes which are allied to the hunting, fishing, and pasto- 
ral states. It will also be found that those artisan castes, which represent or are 
coeval with the metallurgic age, hold almost the same rank as the agricultural castes 
proper. Such a coincidence cannot have been fortuitous ; and the theory enunciated 
in this paper, that the social status of a caste is determined by the stage of culture to 
which it is most nearly allied, shows how it came to pass. 

44. The artisan castes, then, have been classified in this paper under two main 
headings— A, those which preceded the age of metallurgy ; B, those that represent 
this age or are coeyal with it. Details of the castes belonging to each class are shown 
in the accompanying tables, called respectively group 111 A and group III B. Any 
and every kind of commodity produced by human contrivance or human skill, as 
distinct from the raw materials furnished by the soil or from the animals which live 
thereon, is considered to be a product of art and tomes within the functions of the 
artisan castes. This is the sense in which the term " artisan " is meant to be under- 
stood in this paper. This, too, is what the natives of India mean by the term karigar, 
"artificer or skilled workman.'* 

45. A — Preceding the age of metallurgy.— A mong the artisan castes of the lower or 

pre-metallurgic age there is a graduated scale of dignity or rank similar to what has 

been shown to exist in the case of the hunting, fishing, pastoral, and agricultural castes 

already described. In the following list, as in those given above, an attempt has been 

made to groXip them approximately in the order of dignity, the lower being placed 

first and the higher last : — » 

T ( Bansphor-- Basket-maker, &e. 
*' \ Bari— Leaf. plate- maker. 

TT f Chamar— Hide-skinner and tanner* 

1L \ Mochi— Cobbler. 
ttt f r>hnna— Cotton-carder. . 
lil# t Kori— Weaver. 

{Teli— OU-preeaer. 
Ralwar— Spirit-distiller. 
Kumhar— Potter. 
Lunia— Salt-maker. 

All of the industries included in the above list are compatible with a low stage of 
culture, and their equivalents (as might be easily proved if space permitted) may 
be found at the present day among the arts practised by savage and barbarous races 
in various parts of the world. If the people of whom such castes are composed could 
be eliminated from the rest of the Indian population and transported to some, distant 
land or island of their own, any traveller visiting them and recording their customs 
and notions would not hesitate to class them with such races as may be found at the 
present time in Polynesia or in Central Africa, or in parts of the American con- 

46. The names by which the above castes are distinguished from the rest of 
the community show clearly enough that it was some speciality of function, and not 
community, of blood or community of religion, which brought the several castes into 
existence. Thus the name Bansphor, the caste of basket-makers, &c, is derived from 
ban$ 9 " a bamboo," and phur or sphur, " to split 9 ' Another and a better-known name 

Digitized by 


{ *1 ) 

fcr this caste is Dhark&r, which signifies rope-maker, being derived from ett«r, 
"rope," and kdr } "maker." B4ri, the name of the leaf-plate-making caste, is 
derived from bdri, a plantation or orchard z another name for this caste is Bargi, 
derived from barg, a Persian word for leaf. Cham&r, the name of the tanner caste, 
is derived from chamra, " hide." fllochi, the name of the cobbler caste, is derived 
from mot Ana, " tweezers or pincers," one of the instruments used in working leather: 
another name of this caste, "Dabgar,*' is derived from dab, the pressing or flatten- 
ing of the hide. Dhuna, the name of the cotton-carding caste, is derived fromdAtm, 
u to beat out, comb, or card." Bunkar and Joria, synonyms of Kori, the weaver 
caste, are derived respectively from bun, to weave, and jor to join, that is, to put the 
threads together in alternate pairs as Warp and Woof. Teli, the name of the oil-man 
caste, is derived from Xil, the sesame plant, from which oil is very largely extracted. 
Kalwar is probably a variant of Khairwar, a tribe noted for the extraction of an 
intoxicating juice from the khair tree. Kumhir, the name of the potter caste, is an 
abbreviation of kumbakdr, and is derived from kumba, an earthen pot, and kdr, 
maker: the Persian name of this caste is Kashgar, wh|ch is derived from 
kdsahy " a cup," and gar, Ci maker." Lunia, the name of the salt-making caste, w 
derived from laun or naun } u salt," and hence another name for the caste is " Nunera." 
Thus there is not a single caste in this list which is not designated by a functional 
as distinct from a tribal name. 

47. The Dharkar or Bansphor is the lowest caste in the list And this tallies 
with the fact (which could be easily proved from the known characteristics of modern 
savages elsewhere) that the special industry represented by this caste, viz., that of 
making baskets, reed-mats, grass-ropes, <fcc , is the lowest of human arts. The tradi- 
tions of the Dharkar connect him very distinctly with the savage Dom or Chand&l^ 
of which tribe he has sometimes, bat erroneously, been called a clan. The Dharkar 
is simply a reformed Dom ; that is, he has left off eating dogs, burning corpses, 
executing criminals, and sweeping away filth for hire ; but he has retained the ances- 
tral art or industry of making chairs, mats, baskets, &c, out of reeds and cane. The 
differences between them have now become so marked, that the offshoot constitutes 
a stationary or settled caste, while the ancestor remains to this day a wandering 
savage. If the theory with which we started is correct — that the casteless and 
savage tribes supplied the raw materials out of which the entire caste system 
was gradually fashioned— it is quite natural that the lowest of the artisan castes 
should trace its origin to one of the lowest of the indigenous tribes. It is well 
known that basket-making in Europe is the special industry of gypsies. 

48, The Bari stands many degrees higher in the social scale than the Dharkar, 
although the industry which he represents, viz., that of making cups and plates of 
leaves skewered or stitched together with leaf fibres, is almost or quite as low as that 
of the other. This apparent exception to the general theory enunciated in this paper 
is easily explained. If leaf plates and cups were used only by low-caste people, the 
Bari would stand no higher than the Dharkar. But they are used by the highest 
castes as much as by the lowest ; and hence considerations of practical convenience 
have compelled the former to recognize the Bari as pure and respectable enough to 
make plates for them. The same considerations, as we have already shown, raised the 
Kahar or water- carrying castes above the hunting. The result has been that Baris, 
like Kahars, have taken largely to domestic service, in which they are chiefly employed 
as table-servants, waiters, Ac When the master goes out in his palanquin at night, 
the servant accompanies him as torch-bearer, and hence torch-making has become 
one of the special industries of the Bari caste. The strictest Brahmans — those at least 
who aspire to imitate the self-denying life of the ancient Indian hermit— never eat 
off any plates other than those made of leaves, and-it is very probable that, through 
long attendance on such masters, some Baris have imperceptibly learnt the Brahman's 
craft and raised themselves into the ranks of Brahmanhood. Notwithstanding the 
oomparative respect in which the Bari is thus held, he is merely an offshoot from the 


Digitized by 


( it ) 

semi-savage tribes known as Banmanush and Musarha. He still associates wftfr 
them at times ; and if the demand for plates and cups (owing to some temporary 
cause, such as a local fair or an unusual multitude of marriages) happens to become 
barger than he can at once supply, he gets them secretly made by his ruder kinsfolk,. 
and retails them at a higher rate, passing them off as- his own- production. 

49. Descending once more into a social level almost as low as that of the Dhar- 
kar, we come to the caste of Chama>, the speciality of whose function is- that of hide- 
skinning and the conversion of . the hide into leather. The Chamar is a very numer- 
ous caste, and may have sprung out of several different tribes, like the Dom r 
Kanjar, Habura, Cheru, &c , the last remains of whom are still outside the pale of 
Hindu society. Originally he appears to have been an impressed labourer or begar r 
who was made te hold the plough for his master, and received in return space for 
building his mud-hovel uear the village, a fixed allowance of grain for every working 
day, the free use of wood and grass on the village lands* and the skins and bodies- 
of all the animals that died. This is very much the status of the Chamar at the 
present day. He is still the field-slave, the grass-cutter, the remover of dead animals, 
the hide-skinner, and the carrion-eater of the Indian village. The rdpi or tanner's- 
knife is worshipped by this caste at the annual festivalof Diwali. The women are th* 
midwives of the Hinda community. The position of Chamars has improved in some 
cases under British role. Some have learnt the English language, attended dis- 
pensaries, and become native doctors— a profession well-suited to their traditions,, 
as they do not share in the objections felt by some of the higher castes to the use of' 
the dissecting knife. There are a few Muhammadan concerts from the Cham&r caste,, 
who are called Kharkata* or Charkati 

50. The caste of Moohi or cobbler is an offshoot from that of Chamar, but yet 
as castes they have become quite distinct ^ for they do not eat together nor inter- 
marry. The industry of tanning is preparatory to and lower than that of cobblery ^ 
and hence) by the general law announced throughout this paper, the caste of Chamar 
ranks decidedly below that of Mochi. The ordinary Hindu does not dorisider.the- 
touch of a Mochi so impure as that of a Chamar, and there is a Hindu proverb to the 
effect that " dried or prepared hide is the same thing as cloth ;" whereas the touck 
of the raw hide, before it has been tanned by the Chamar, is considered a pollution. 
The Mochi does not eat carrion like the Chamar, nor does he eat swine's flesh ; nor does 
his wife ever practise the much-loathed art of midwifery. He makes the shoes 
leather aprons, leather buckets, harness, portmanteaux, &a, used by the people of 
India. As a rule, be is much better off than- the Chamar, and this circumstance has 
helped amongst others to raise him in the social scale. A considerable proportion 
of the Mochi caste has become Muhammadan. 

51. The two next caster are those whose special industry is- connected with 
cotton. The first, the I>huna, called also- Behna, 4s the cotton-carder and thread- 
twister. The second, the Kori, Joria, or Bunkar, is the weaver or cloth-maker. As 
the former industry is preparatory to and lower than the latter, so the caste of Dhuna 
ranks below that of Kori* 

The material used by the Dhumvis furnished by the Salmali or cotton -bearing 
tree, which is, ho doubt, identical with what Herodotus called " the wool-bearing 
trees^of Ethiopia." The Dhuna gather* the ootton-pod as it falls, beats out its 
contents into fine particles with an instrument resembling a bow, and twists them 
into threads. The industry is closely akin to that of making ropes out of the fibres 
of grass, reeds, and roots,.,in which the Dharkar excels. But in consequence of its 
being more refined, the cast of Dhuna stands. higher than that of Dharkar. 

52. The material used by the Kori is the thread supplied by the Dhuna ; and 
thus the weaver caste has risen imperceptibly out of that of the cotton-carder in. 
the same way as the cobbler caste has risen out of the tanner. The art of weaving 
or plaiting threads is very, much the same process as that of plaiting osiers, reeds*, 

Digitized by 


( 23 ) 

and grass, and converting them into baskets and mats. This circumstance explains 
the puzzle why the weaver caste in India stands at such a low social level. la 
countries like England, which have gone beyond the agricultural stage and entered 
fully into the scientific and commercial, weaving (by machinery) has become a very 
respectable form of industry. Bnt India has never gone beyond the agricultural 
stage; and hence weaving has remained, what it originally was, the occupation of an 
ignorant and inferior class of people. The Kori ranks, however, several degrees 
above the Cham&r or tanner ; for amongst Hindus herbs and their products (cotton 
being of course included) are invariably considered pure, while the hides of dead 
animals are regarded as a pollution. Owing to the large importation of machine- 
made stuffs from Europe, the business of the Kori, whose weaving is all done by hand,, 
has become very much depressed, and many of the rising generation are growing up 
totally ignorant of the ancestral craft. They are now taking to grass-cutting, plough- 
ing, &c., like Chamkrs. The weaving caste ha9 two functional names (Joria and 
Bunkar, the meaning of which has been already explained), and one tribal name, Kori. 
The last shows that the Kol (or Kor) tribe contributed largely to the personnel of the 
taste ; and this fact is quite in keeping with our theory, that weaving represents a low 
•stage of art, and that the first weavers in India sprang from a wandering and semi- 
savage tribe, such as the Kols still are. Dluhammadan convert* from the weaver 
caste are called Jul&hA* 

53/ The four castes remaining to be described in this list stand at rather a higher 
level than any of thdse previously named, except the B&ri, whose case is peculiar. 
These four are allowed, but not encouraged, to enter a Hindu temple, whereas all the 
preceding castes, except the B&ri, are absolutely excluded. 

The Teli is perhaps the lowest of the four. He presses oil from the HI or sesame 
plant, and such oH is largely used by all classes of the Indian community for cooking 
food, anointing the body, softening leather, and healing sores* He also extracts oil 
from the seed of the ninfe tree, of the castor plant, the mustard plant, and the thorny 
shrub called Bharbh&nd, which grows wild in the jungles. The art of the oilman has 
been greatly improved 1 by that of the agriculturist, who has supplied him with many 
kinds of oil-yielding plants which the jungle does not afford. Nevertheless, the art 
itself in its^ earlier and less refined forms preceded the age of agriculture, and must be 
counted among the lower types of industry. This is why the caste has always ranked 
low. The Teli never extracts oil from the bodies of animals, as Kanjars and others 
of the lowest tribes of Hindustan are accustomed to do. Probably, however, he 
learnt the first rudiments of his art from such tribes, and is himself- an offshoot from 
. them. 

54. The Kalwar ranks a little higher than the Teli, because there is more skill 
and less dirt in the practice of his art. His trade as a private occupation has been 
destroyed by the British Government, which has taken the distilling and sale of 
liquors entirely into its ownhands. The Kalwdr still finds some opportunity, however, 
of following his old pursuit by working in the Government distilleries and taking 
out licenses for the sale of spirits. But the majority have taken to* other kinds of 
trade or to agriculture, the common goal to which all the decayed industries of India 
are tending. The art of the Kalw&r, like that of the Teli, has been known to almost 
all the backward raceskof the world, and cannot by any means be counted among the 
higher types of industry. Hence the status of the Kalw&r ha* always been low. The- 
name, as I stated abov^, is probably a variant of Kl&irwdr, or " catechu -maker,'' a 
process which is very similar to that of drawing juice from the palm-tree and fer- 
menting it into a spirit. The name of Kalw&r then, if this identification is correct, 
implies that the caste has sprung up out of such tribes as Chain,. KhairwAr, Musarha,. 
Ac, all of whom are skilled in making the intoxicating juice called catechu. 

55. The Kumhar or potter represents a form of industry about equal in rank, 
to. that of the Teli or Kalw&r, and hence his status in the social scale is much the* 

Digitized by 


( u J 

same as theirs. It is now a well-established fact that pottery was known in the Stone 
age of the world ; and the art was found to be in fall rogue among the Fijian savages 
of the Pacific and among the native races of North America, when they were first 
visited by voyagers from Europe. The potter's wheel is an application of the mill- 
stone, but the lowest stage of the art can probably be traced to basket-making. Men 
discovered without much difficulty that their baskets could be made water-tight by 
smearing them over with clay. Afterwards by putting such baskets on the fire, it was 
found that while the reeds and osiers were consumed, the clay not only remained, but 
was strengthened and hardened by the fire.' Eventually clay vessels were made 
without the help of the basket framework, and hence pottery was developed into a 
distinct art. All the casteless tribes of India are skilled in making baskets; and 
hence we may infer that the potter caste sprung out of one or more of these tribes, 
and that the bharkar may have been the intermediate caste. 

56. The Lunia or salt-making caste closes this list. His industry has been even 
more destroyed than that of the Kalw&r, and by the same cause. Now he has taken 
to road-making, well-digging, mud-masonry, Ac., and his industry corresponds very 
nearly with that of the English navvy. In the days of his prosperity he manufactured % 
salt and saltpetre out of the usar or alkaline soils, in which Upper India abounds. He 
still does so occasionally at night time and in secret As a digger and delver in the 
earth he not unfrequently stumbles upon field-rats, and these form part of his diet, 
when he can get them. His status in the social scale is about equal to that of the 
Lodh or Lodha, the lowest of the agricultural castes. 

57.- In connection with the above castes, some reference must be made in this 
place to the position which some of them hold in the constitution of the Hindu town- 
ship or village commune— «a kind of municipal and self-sufficing body, which charac- 
terized the civilization of the Hindus from very early times, and which is not by any 
means extinct. Allusion has already been made in para. 15 to the village watch- 
man—a post* assigned from the oldest tim&s to some man of the hunting or predatory 
tribes (such as P&si, Khang&r, Arakh, Bhar, <fec.) who abstained from robbing, and pre- 
vented others from robbing, the residents of the village, on condition that he received 
bis stipulated reward and was allowed to build a hut for himself and family on the out- 
skirts of the hamldt. Among the artisan castes which we have just described there 
' are two — the Chamfir and Kumhir — who are essential factors in the constitution 
of the township. The Cham&r removes the carcasses of cows and buffaloes, takes 
off their hides, tans them into leather, makes the leather buckets used for drawing 
water from the wells into the fields, and makes shoes for some of the residents. 
Were it not for the leather buckets made by Cham&rs, no water could be drawn from 
wells for the irrigation of crops during the long dry mouths of the year, and the labour 
of the husbandman would be lost. The function of Kumh&r or potter is useful &t- all 
times, but comes more markedly into use at times of chiid-birth, marriage, and fune- 
rals. On the twelfth day after a child has been born, when the final bath is given to 
the mother on leaving her sick chamber, all the earthen pots of the household are 
discarded and replaced by new ones. At the time of the celebration of a marriage the 
father supplies the new jars, in which Ganges water, barley, and lighted oil are placed 
in honor of Ganesha, the god of luck, and without which the marriage ceremony could 
not be performed. On the tenth or thirteenth day after a death, all the -earthen vessels 
of the house are destroyed and replaced by new ones, as on the twelfth day after a birth. 
Of each of the three functionaries whom We have named— the watchman, th£ hide- 
tanner, and the potter— it may be said that on special occasions, or for services of 
exceptional merit, they receive some special remuneration in cash or kind ; and thai 
for ordinary services they receive a stipulated portion of the produce of each of the 
two annual harvests. Additional presents are made to them in the great anniversa- 
ries of Diw&li, Holi, and Dashara. 

58. It remains for me to describe in what respect the names and figures shown 
in group II I A, differ from those given in the census report; — 

Digitized by 

, y Google 

t ** ) 

(a) Basket-maker*.— Kharot, Gogh, and Charu, against whom 3,610, 369, and 
514 persons have been recorded respectively in the census report, have been included 
in my own table amongst Dharkars or Bansphors, to whom they really belong. For 
the name Banspbor given in the census I have substituted Dharkar, as the former 
name is sometimes applied to men of any tribe or low caste (such as that of sweeper 
for example) which follows the oocupation of basket-making, &c, as an accessory to 
other callings. 

(6) Leaf~plate-makers.—Ba,rgi t Bargahi, and Sijwari, against whom 1,189, 3,777, 
and 376 persons have been reoorded respectively in the census report, are not separate 
castes, but merely synomyms of Bari, and have therefore been included in this caste 
in my own table. 

(e) Leather-workers. — No mention is made in the census report of the Moohi or 
cobbler caste, the members of which have evidently been mixed up with Chamars. I 
am unable, therefore, to separate the figures, but I have added the name Moohi to 
the list of castes belonging to this table. 

(d) Kharkatas are Chamars who have become Muhammadan j they have, 
therefore, been included by me amongst Chamars. 

(e) Dabgar, Karal, and Jaiswar, against which 1,231, 333, and 832 persons have 
been recorded respectively in the census report, are not separate castes, but sub-classes 
or synomyms of Mochi, and under this name 1 have included them in preparing my 
own table. 

(/) Thread-twisters.— *Katwa, against which 122 names have been recorded in 
the census report, is not a caste, but a sub-class or synonym of Dhuna, under which 
name I have included them* 

(y) tifatftfr.— Joria, Bunkar, and Balai, against whom 10,923, 6,635, and 189 
persons have been respectively recorded, are not separate castes, but synonyms of 
Kori, under which name I have included them. 

(h) Oil-makers.— Badiphul, against which 429 persons have been recorded, is 
included by me amongst Telis. 

(t) Potters. — Balahi, against which 121 persons have been recorded, is included 
by me amongst Kumhars. 

[j) Raj and Kathyara, against which 9,683 and 295 persons have been recorded, 
are not names of castes, but of the occupation of brick-layers, As such men are 
generally of the potter caste, I have included them amongst Kumhars. 

(k) Salt-makers.— Parbia, Agaria, and Sunkar, against which 6,205, 1,384, and 
1j084 persons have been reoorded, are sub-classes or local titles of the caste of Lunia, 
under which name 1 have therefore included them. 

59 # J. — Coeval with the age of metallurgy.— The upper artisan castes (vide group 
IIIB.) are those which represent, or are coeval with, the age of metallurgy. They 
may be arranged as follows in point of rank or social respectability, the highest being 
mentioned last;— 







... Sangtarash 

c Ramangar and Tirgar 
••. ) Barbai ... 


(Lobar ... 

Kasera ... 

( Sonar 
*** I Tarkihar ... 

iDarzi ~« 

Rangrei ... 

... fialwal ... 


Bow-maker and painter. 








Earing- maker. 






Digitized by 


( M ) 

60. The above castes, as was shown above, represent higher stages of art, and 
consequently stand higher in the social scale than those named in the preceding list 
The marks of social superiority possessed by these castes are the following : — (1) Men 
of the highest Hindu castes or classes can sit down in their company without loss of 
dignity; (2) no Hindu, not even a Brahman, would refuse to drink water drawn by 
their hands; (3) they are more punctilious than the castes named in the previous list 
in the Hindu observances of bathing, praying, Ac., at the appointed times and in the 
appointed methods ; (4) they are allowed the freeest access to the inside of a Hindu 
temple, whereas the previous castes are either discouraged from entering a temple or 
absolutely excluded ; (5) they decline to drink water drawn by any caste whose ser- 
vices as a water-carrier would be rejected by a Brahman; (6) they invariably eat 
their food on the chauka or floor on which it baa been cooked, and would reject as 
impure any cooked food that has been taken outside the chapka. On the other hand 
they rank decidedly below Brahmans from the fact that they allow the marriage of 
widows and indulge in the use of spirituous liquors ; and it is needless to add that the 
functions which they represent, being of a purely material or industrial order, are of 
much less social dignity than that of Brahman or priest. The average rank of the 
castes is about equal to that of Tambolis and Malis (see above para. 33 », but 
below that of Tagas and Bhunihan, since neither of the two last named allow the 
remarriage of widows. 

61. As in most of the preceding lists, the names by which the several castes 
&re called show very clearly the nature of the functions yrhich distinguish them from 
the rest of the community, and by which they have been called into existence as sepa- 
rate castes or classes. Sangtarash means "stone-trimmer" or " stone-cutter, 1 * 
being derived from two Persian words, sang and tarash. Kam&ngar and Tirgar mean 
respectively "bow-maker 11 and "arrow-maker," being derived from kamdn (bow) 
and tir (arrow), both of Persian origin. Barhai (oarpenter) is an abridged pronunci- 
ation of bardhai and is derived from the Sanskrit root bardh, which signifies " to bore 
a hole." Gokain or Qokai (wood-carver or engraver) is derived from the Hindi 
khonch, which means to cut a hole or scoop out. Lohar is derived from loha, iron; 
Kasera from Icdruta, bell-metal ; Thathej-a from thathi, a brass platter ; and Sonar from 
savarna or sona, gold, all of Sanskrit origin. Manihar and Turkihar, the names of 
the ornament-making castes, ape derived respectively from tnani, a gem, and turki, an 
earring, both of Sanskrit origin. Darzi, the name of the tailor oaste, is derived from 
Persian darz, sewing, or darzan, a needle, the Persian name having superseded the 
old Hindi name, "suji," which was derived from a Sanskrit root signifying "to sew/' 
Patwa, the fringe-making caste, and Chhipi, the cloth-stamper, come respectively from 
pat, a piece of silk, and c/lAapna, to print, both of Sanskrit origin. Rangrez, the name 
of the dyer caste, is derived from two Persian words, rang, colour, and ret, sprinkl- 
ing. Hal wai, the name of the confectioner caste, is derived from haltca, a kind of 
sweetmeat in very common use among the middle and upper classes of the community. 
Thus every name in this list, without exception, is functional and not tribal. 

62. The caste first named in the list is Sangtarash. But I feel much doobt as 
to whether " caste " is the right term to apply to men of this class. Jt is certainly 
the name of an hereditary occupation, that of stone-outting ; and it is certain that 
many men, on being questioned what their jat or caste is, wil| answer " Sangtarash." 
But if they are pressed with further enquiries regarding their origin, Ac., it will gene* 
rally be found that there is some other caste with which they still own connection, and 
to whose limits they arp still restricted in the matters of marriage and convivial inter- 
course. Until it can be proved that "Sangtarash" represents a certain class of the 
comm unity which disowns connection with any other c^iste and can neither intermarry 
nor eat and drink with any but persons of their own fraternity, it caunot be said 
that any such caste as Sangtarash exists ; for this is what the word " caste " implies, 

The men t* ho follow the occupation of stone-cutting are generally Khatiks, Lunias, 
Ahirs, Gokais, Gosh^y^ns, apd Brahuians. If there is a paptjB of Sangtarash in tip tru* 

Digitized by 


{ 27 ) 

sense of the word, it will probably be found that it is made up of men who have 
deserted the ranks of the lower of the castes just named, And have thus established a 
new caste or trade union of their own. Goshayens and Brahmans may and do follow 
the occupation of stone-cutting, but they would never consent to leave their own 
castes in order to enter a lower one. The Persian etymology of the name SangtarasK 
implies that the caste, if it exists at all, is of modern origin and foreign to the old 
Hindu system. The caste so called seems to have been continually struggling into 
existence, but never to have quite come to the birth; and this state of incompleteness 
has been due partly to the incongruous character of the men who have taken to the 
calling, but chiefly to the 3carcity of stone and the consequent fewness of men to whom 
such an occupation is open. In Upper India one of the chief centres of the stone* 
cotter's art is Jduttra, but not (1 am told) so important a centre as Benares or Mirza- 
pur. The art has been developed chiefly in connection with the making of idols and 
the ornamentation of temples, and this is why it has been cultivated by Gosh&yens 
and Brahmans. In the JJenares and Mir^apur districts it is not uncommon to find the 
door-posts of private houses made of stone instead of wood, and such door-posts are 
engraved with figures of Ganesha, the god of luck, of Lakshmi, the goddess of wealth, 
*nd of a fish as the incarnation of Vishnu. The n*en who work the quarries of Mirza- 
pur are generally Ahirs or cowherds, who from grazing their cattle in the same neigh* 
bpurhood have learnt to cut stope out of the rocjc and carry it into the towns for sale. 
Such Ahirs are sopjetinjes called Peshraj— that is> njen who bring and prepare stones 
for the mason. 

63. From the workers in stone we come to the workers in wood, and three 
castes can be included in this category. Of wood-workers generally as of those who 
work instone it should be understood that the germs of their respective arts are coeval 
with the savage age and are among the very earliest contrivances of human industry. 
But it is only when iron tools have come into vogue that either of these arts can 
acquire that degree of refinement which they have reached in India, and hence it 
is correct to include them, &s has been done in this place, among the arrs coeval with 
the age of metallurgy. The castes which practise them are in point of social status 
*bout on the same level as those which practise metallurgy itself. 

Kamfing^r and Tirgar, wbicfy literally means the makers of bows and arrows, are 
sub-classes or offshoots of a single tribe or c*ste which in early Hindu times was 
Jcnown as PhAnuk, and which manufactured the spears, clubs, bows and arrows used by 
the hunting pnd pastoral tribes. The material cjiiefly used then, as jiow, for suclj 
purposes was tfre bamboo. Some remains of this Dh$nuk or bow-making caste are 
ptill to be found in the city of Benares, where they sell bpws, &c, as {children's toys, 
In ancient India the bow was esteemed the best and greatest of weapons ; and many 
of the old Kshatriya warriors are described as making their own bows and addressing 
them by names as if they were animated beipgs, like the Sikh doing worship to his 
guns. At Jha present day, when bows and arrows ape no longer used for serious 
purposes, they are made only for toys ; but as the demand for such articles is small, 
Kamangars hare taken to many other kinds of industry. The use of the lathe, in which 
they specially excel, has put them in the way of turning out much finer work than the 
old bow or spear of bamboo. With this instrument they make toys and ornaments of 
all kinds, and they have acquired the additional art of painting them with coloured 
lacquer. From colouring rounded surfaces, they have learnt to paint flat or uneven 
ones, such as doors, window -frames, palanquins, &c., and thus it has come to pass 
that painting, rather than bow-making, is now. the speciality of jLhe Kwdngar 
caste. The beautiful painting which is laid on pottery ware is chiefly done by theip* 
The old fpwling and hunting tribe of Dh4nuk is still numerous in Upper India, bqt 
the bow-making portion of this tribe has for the most part abandoned the name of 
Dbfinuk, and assumed that of Kam&ngar and become Muhammadan. AU the Kamin- 
gars and Tirgars that I have seen or heard of profess the creed of Islam, and woulrf 
jtiojt alio* themselves to be mentioned in the same bjrcath witfc the despised caste pf 

Digitized by 


( » ) 

Dh'Anuk. The tradition of their origin has sutvfved, however, firstly, in the name 
Kam&ngar or bow-maker, which is now almost a misnomer ; secondly, in the fact that 
Tirgar women are sometimes employed as mid wives in Mubammadan households, just 
as Dh&nuk women are employed in Hindu ones ; and, thirdly, in tbe fact that Kam&n- 
gars are sometimes employed as bone-setters by Muhammadans, as Dh&nuks are by the* 
lower castes of Hindus. In reference to this third link it should be explained that 
Dhanuks and Kam&ngars are both skilled in straightening bamboo poles which are 
crooked or in curving the handles to cane walking-sticks. The first process in curving 
the canes consists in steeping them in boiling oil. Similarly the first process towards 
setting a bone is by softening the limb with heated oil ; and thus the two industries 
have become amalgamated. 

64. Tbe Barhai is the well-known and time-honoured carpenter of the old Hindu 
township. As such he is a kind of public servant, and no village would be complete 
which did not contain one or more of such functionaries within its* circle. Regularly 
once a year, in tbe month of Asarh (which takes in the latter half of June and the first 
half of July), and on some auspicious day selected by the Brahman or village priest, 
the agriculturist takes his plough to the carpenter to be repaired for tbe year's work, 
and rewards him partly in cash and partly with a gift of sweetmeats. All future- 
repairs, until the next As&rb comes round, are done gratis, except that at the close of 
each harvest the carpenter receives a stipulated proportion or the produce. He i* 
accustomed, too (like tbe watchman, the hide-tanner, and the potter described in 
para. 57), to receive presents of grain at tbe three great festivals of the Dashara, 
the Diff&li, and tbe Holi. His status is about equal in rank to that of the Kurmi 
or peasant, with whom his own interests are so closely linked. Besides making the; 
plough used by husbandmen, be constructs the wagons, boats, boxes, bedsteads, Ac, 
used by the community at large. There is reason to believe that the caste of carpenter 
sprang out of the fishing rather than the hunting tribes. The single-logged canoe* 
which » to be found among tbe most backward race* in various parts of the world, 
and was known to tbe ancient Britons prior to the Roman invasion, is still made m 
India by the Barhai caste ; but the first men who made boats of this rough description 
must have been themselves fishermen. 

65. Tbe speciality of the Qokain as distinct from that of the Barhai consists in 
carving and engraving the materials which tbe Barhai has furnished. His art, then is 
more refined than that of the Barhai, and hence his social status i& somewhat higher. 
There can bo little doubt that the Gokain caste ha» sprung out of the Barhai in the 
same way as tbe Barhai himself sprang out of tbe Kewat or fisherman. The carvings 
on boats, door-posts, bullock-carts, &c, are the work of this caste. When the demand 
for such labour is slack, he takes to using the lathe like the Kamingar, and paints with 
various coloured lacquer the toys and ornaments which he turns out for sale. The Gokain 
is also noted as a carver and engraver of stone, especially of the ornamental stone 
pillars used at the doorways of temples and private houses. In this kind of art he is the 
most skilful of those castes which sometimes allow themselves to be called by the 
generic title of Sangtarash. 

66* From the workers in stone and wood we come to the important series of 
castes, whose special industry consists in the working of metals, and who are 
therefore the typical representatives of the metallurgic age. The order in which I 
will now describe them coincides approximately with that of theif rank or status in 
point of social respectability. 

The Loh&r or iron smith holds a place in the constitution of the village com- 
mune exactly similar to that of the Barhai, and be is remunerated on the same 
principle. The work of tbe wood-cutter and ploughman could not be earned on 
without him, and to the general public he is the maker of all kinds of iron tools 
the metal-polisher, edge-sharpener, farrier, knife-maker, Ac. It is owing to the 
ubiquitous industry of the Lohar that the stone-knives, arrow-heads, and hatchets 
; of tiie indigenous tribes of Upper India (even those of the Dom, Kanjar, Ac., who in 

Digitized by 


( » ) 

character and habits are still almost as savage as their ancestors were some six 
thousand years ago), hare been so entirely superseded by iron ones. The memory of 
the Stone age has not survived even in tradition* In consequence of the evil associa- 
tions which Hinduism has attached to the colour of black, the caste of Loh&r has not 
been able to raise itself to the same social level as the three metallurgic castes which 
follow. A piece of iron is not unfrequently kept in Hindu houses, especially in the 
lying-in chamber, to avert the evil eye. 

67. The three castes which follow have no place in the staff of the village 
community similar to that held by the Loh&r and Barhai. But the two brazier castes, 
the Easera and Thathera, are held in almost universal request. There are scarcely 
any articles of furniture which a Hindu prizes so highly as the brass lota with 
which he draws water from the well, the batud or brass bowl in which he cooks his 
food, the thdthi or brass platter on which he eats it, and the tdm or copper platter 
on which he places the offerings of ghi, rice, water, Ac, intended for the propitiation 
of his patron deity or deities. It is a mark of the deepest social depression not to 
be the owner of such artioles, and no expression' conveys a stronger idea of poverty 
than to say of a man that " he does not possess even a lota." In times of famine, 
it is by pawning or selling his lota that a man makes his last attempt to save, himself 
and family from starving. The Ka*era's speciality lies in mixing the softer metals 
(zinc, copper, and tin) and jnoulding the alloy into various shapes, such as cups, 
bowls, plates, &c. The Thathera'a art consists in polishing and engraving the 
utensils which the Kasera supplies. His work, then, is more delicate than that of the 
Kasera, and his status in society is raised accordingly. The Thathera is thus re- 
lated to the Kasera in the same way as the Gokain is related to the Barhai, the 
Mochi to the Cham&r, the Kori to the Dhuna, or the Dhuna to the Eanjar. The 
upper artisan castes have been formed out of those standing immediately below them, 
just as the lower artisan castes have sprung out, of the savage tribes. 

68. The Son&r's work lies in gold and silver. He stands a little higher than 
those castes who manipulate the inferior metals, because his art, being more delicate 
and costly than theirs, brings him more into contact with the higher and wealthier 
•lasses of the community. It is not every Son&r who can set up business with 
his own capital : hence it is very common for men of the Ehattri or banking caste 
to take a staff of Son&rs into their employment and set up a jeweller's shop. 
Owing to the dishonesty of a few of its members, the Son&r caste has in some places 
acquired rather a bad reputation for receiving stolen jewellery and throwing it into 
the crucible, so as to render recognition impossible. It is said, too, that there is a 
secret language in which this kind of traffic is. carried on between themselves and 
their accomplices. This is probably one of the offences alluded to in Mann's 
Criminal Code in the following words : — " The most pernicious of all deceivers is a 
gold-smith who commits frauds ; the king shall order him to be cut piecemeal with 
razors" (IX, 292). In contrast with the evil significance attaching to the black 
metal, iron, the brightest associations have been fastened by Hindus upon the shining 
metals, gold and silver : and this fact has tended to raise the Son&r caste in propor- 
tion, notwithstanding their suspected dishonesty. It was the belief of ancient Hindus 
that gold and silver, the most perfect of metals, arose out of the combination of fire 
with water, the purest of elements (Manu, V, 113). 

69. The two castes now to be named (Blanihar and Tarkihar) are those 
whose speciality lies in miking and moulding glass— a pro I net which could only 
have come into vogue after the rise of the art of metallurgy ; for both processes 
involve the use of the same element — firta— -for melting, mixing, and transmuting the 
materials furnished by nature. The two glass-making castes stand on about the 
same social level as the Son&r. One industry — the making of mirrors — is common 
to both. Glass windows appear to have been unknown in India until Europeans 
introduced them. But from allusions occurring in the early literature, it is certain 


Digitized by 


( 30 ) 

that the ancient Hindus were acquainted both with looking-glasses and with the 
glass lens by which fire is drawn from the sunbeams. At the present day both 
castes exist almost entirely for the convenience of women. 

The Manih&r is not a member of the staff of which the old Hindu township is 
composed to the same extent as the ironsmith or the carpenter. But if some 
Tillages here and • there happen to have no] Manih&r living in their midst, there is 
certain to be one or more in some other village within easy reach. His special indus- 
try is the making of glass-bracelets or bangles. The domestic life of a Hindu woman 
would be almost impossible without this : for the glass bangle is not worn for per- 
sonal ornament, but as the badge of the matrimonial state, like the wedding-ring 
worn by women in Europe. The girl puts on her glass bangle for the first time 
when she starts on her bardt, the marriage procession which leads her away to the 
bridegroom's house. After child-birth the mother, on leaving the room in which she 
has been lying, breaks her bracelet and is immediately provided with a new one by 
the wife of the Manih&r, who receives cash in exchange for the bracelet and a gift of 
sweetmeats or grain for the ceremony of putting it on. When the husband dies, the 
widow must, at the moment of his death, break the bracelet off her wrist in token 
of widowhood, and till the day of her] death she must never wear another, unless 
she marries again. These observances and prohibitions are binding on all olasses of 
Hindus, even the very lowest ; and henoe it is no wonder that the Manih&r 
commands general respect and is very widely scattered among the community. The 
word current amongst Hindus, for bracelet in general is churi ; but such is the value 
attached to the glass bangle worn as the badge of matrimony that the caste by whom 
it is made has received the metaphorical title of " Manib&r," which means literally 
" a maker of gems or jewels." Jewellery, in our sense of the word is made only 
by Sonars. Any articles other than the glass bangle may be taken off or put on at 
pleasure and are intended for ornament only, but the glass bangle must be worn day 
and night, so long as the husband is alive. 

The Turkih&r caste makes the glass beads with which earrings and necklaces are 
studded. The ground-work of these ornaments is generally a piece of palm-leaf 
variously shaped and coloured. The earring worn by Hindu women is not a curved 
ornament attached to the ear by a hook, but a straight one, the thinner end of which 
is inserted into the lobe of the ear, while the thicker end studded with beads of glass 
is made to stand out It is not considered 6t for a widow to wear such ornaments, 
but the rule is much less strictly observed than that connected with the glass bangle. 

70, The four next castes are those whose special industry lies in the making of 
wearing apparel. The Darzi or tailor makes clothes of silk as well as cotton fabrics, 
and works ohiefly for men. The Patwa or fringe-maker uses the same materials, but 
works chiefly for women. The Chhipi or stamper is employed on cotton stuffs only ; 
the Bangrez or dyer on cotton, silk, and woollen stuffs alike. It is easy to see that 
these four castes belong to the more advanced stages of arts, and that such castes 
could never have come into existence, if they had not been preceded by certain lower 
castes, whose industries were preparatory to theirs* Thus the weaver and the thread- 
twister (Kori and Dhuna), whom webave already described, paved the way for the 
tailor, dyer, &c, and the weaver himself learnt the rudiments of his art from the 
savage who gathered reeds and osiers from the forest and plaited them into baskets 
and snares. It is certainly a fact that the rank which these several castes hold in 
the Hindu sooial scale coincides with the rank of the arts corresponding ; and as the 
same coincidence has been shown to exist in the groups previously named, it must be 
the result of a general law and not of chance The substitution of Persian names 
for the old Hindi names of tailor and dyer indicate, what is the fact, that these 
industries have been largely patronized by Muhammadans, and that many members 
of the castes have embraced the creed of Islam. Silk fabrics, till lately, were not 
made in India, but imported vid Cabul. Woollen stuffs were, and are still, manufac- 

\ [ 


( 31 ) 

tared by the Gadaria or shepherd caste. The cotton fibre, as we have shown already, 
grows wild on the simal tree, is twisted into thread by the Dhuna, and woven into 
cloth by the Eori. 

71. The last caste of artisan to be named in this list is the Halwai, whose special 
craft lies in making sweetmeats, and who is therefore sometimes called Mithiya. 
The materials used in Hindu confectionery are milk, butter, flour, and sugar. It is 
obvious, then, that the art of the Halwai presupposes the industry both of the 
milkman and of the agriculturist. In other respects, too, his art implies rather an ad- 
vanced stage of culture, and hence his rank in the social scale is a high one. There 
is no caste in India which considers itself too pure to eat what a confectioner has made. 
In marriage banquets it is he who supplies a large part of the feast, and at all times 
and seasons the sweetmeat is a favourite viand to a Hindu requiring a temporary re- 
freshment. There is a kind of bread called puri, which contains no sugary element ; 
but yet it is specially prepared by men of the sweetmeat caste. It consists of wheaten 
dough fried in melted butter, and is taken as a substitute for the ehapati or wheaten 
pancake by travellers and others who happen to be unable to have their bread cooked 
at their own fire. With the exception of Brahmans, there is no class of men in India 
whioh declines to eat a buttered pancake prepared by the Halwai ; and considering the 
immense amount of fuss (involving even forfeiture of caste) which is attached to the 
domestic fire-place, this says much for the respect in which the Halwai is held. As in 
the case of the Bari, the caste whioh makes leaf-plates for all classes of the communi- 
ty, considerations of general convenience have no doubt contributed something to 
the social respectability of the confectioner caste. The Halwai is a strict Hindu, but 
he has not gone so far in the direction of Brahmanism as to disallow the re-marriage of 
widows within members of his fraternity. 

72. The names given in the census report do not differ much from those collected 
and arranged in my own table (group 1IIB.) The differences are as follows : — 

(a) Kam&ngar and Tirgar, which are shown as separate castes in the census 
report, and against which 1,365 and 309 persons have been recorded, are rather differ- 
ent names of a single caste. I have therefore included them both under the name 
of Kam&ngar. 

(6) In the census report Nalband or farrier, Barhia or edge- sharpener, Saikal- 
gar or metal-polisher, and Lon or point-maker, are given as separate castes ; and the 
number of persons recorded against them is 3,230, 410, 845, and 209 respectively. 
These, however, are not separate castes, but only different occupations within the same 
caste of Lohar or ironsmith ; 1 have therefore included them all among Loh&rs. There 
is a tendency in these specialities to become hereditary ; but the men who practise 
them have not, so far as I can learn, detached themselves from the parent stem, so as 
to render inter-marriage or the right of eating together impossible. 

(c) Niyaria, or washer of gold-dust, against which name 1,276 persons have 
been recorded in the census report, is scarcely a separate caste, but rather a sub-class 
of Son&r or gold-smith. In this caste, therefore, I have included them. 

73. This brings us to the end of the wealth-producing castes. They consist, as we 
have shown, of two great series or groups ; the first of which comprises all those castes 
whose function it is to call forth and utilize the productive energies of land and water, 
and who are sub-divided into hunting, fishing, pastoral, agricultural, and landlord ; 
while the second comprises those castes who practise the various arts or handicrafts, 
without which the industries of man on land or water could not be carried on, and 
these, as we have shown, are sub-divided into upper and lower, according as they do 
or do not precede the metallurgic age. All of these castes are producers of wealth, 
with the exception of the Ghattri or landlord, who merely protects it ; and he, too, 
is now becoming a producer, because he is learning to handle the plough. Before leav- 
i ng the subject of these castes a few words must be said in this place in reference to 
the modem and widely accepted theory which divides the population of India into two 

Digitized by 


( 32 ) 

great racial sections known as Aryan and non-Ayran. To take the landed industries 
first. There is no reason to donbt that the indigenous tribes of India, before the Aryan 
invader had arrived, were versed in the arts of hunting and fishing and the rearing and 
grazing of cattle, and that they knew how to raise, on the alluvial banks of their rivers, 
crops of rice, juar, bajra, and any other grains that are indigenous to the soil. It is 
very probable, too, that the fishing tribes know how 4o raise crops of the singh&ra or 
waternut in the swamps and marshes in which their country abounds. The indigenous 
tribes, then, had eome well within the nomad state, but were acquainted with agri- 
culture only in its most rudimentary form. What the Aryan tribes did, then, was to 
introduce more regular habits of industry, better kinds of tools, better methods of 
agriculture, and the grains of the temperate zone, such as wheat, barley, oats, gram, 
none of which are indigenous to the Indian soil, and which will only grow during the 
Indian winter, and thus gradually to establish the agricultural state proper, with all 
the settled industries that accompany it, in substitution for the wild and migratory life 
of the aboriginal hunter or nomad. Again, to take the case of the various arts or 
handicrafts. There is no reason to doubt that the lower kinds of art, such as basket- 
making, thread-twisting, weaving, oil-pressing, pottery, salt-making, the brewing of 
spirituous liquors, Ac., were known to the indigenous tribes of India before the Aryan 
stranger had appeared ; for the like have been found to exist among modern savages, 
and the manners and notions of the Indian castes whieh practise these crafts are much 
on the same level as those of the backward races in other parts of the world. What 
the Aryans did, then, was to introduce the art of metallurgy, and with this the higher 
kinds of art, including that of agriculture itself, which invariably come with it or fol- 
low it. Now if we are to suppose that the Indian castes are distinguishable into Aryan 
and non-Aryan, the most rational application qf such a theory would be to place the 
agricultural castes aud the metal-working castes on the Aryan side, and all the castes 
which preceded these on the non- Aryan. But the theory has been made to run in a' 
groove far less consistent with the fact. It accepts as true the old fourfold division 
of castes into Brahman, Kshatriya, Vaisya, and Sudra, and says that the first three are 
of Aryan blood, while it degrades the fourth, in which all the agricultural and metal- 
working castes are included, to the rank of non-Aryan. The only escape from these 
contradictions is to put the Aryan out of account altogether ; that 1*9, to recognize once 
for all that whatever improvements or changes in industry, language, and religion he 
may have introduced some four thousand years ago into the land of his adoption, bis 
blood has become absorbed into that of the indigenous population, and that he no 
longer exists in India as a distinct racial unit, much less as a basis for the explanation 
or classification of caste. 

74. Trading castes.— The remaining castes, that is, those which are not connect- 
ed with any industry either by land or water and are not engaged in any kind of 
art or handicraft, and which are, therefore, distributors and consumers, but not pro- 
ducers, of wealth, may be classified under three main headings as (1) trading, (2) serv- 
ing or professional, (3) priestly. These three together make up only about one 
quarter of the entire population of the province ; and this is quite as large a propor-, 
tion as could be expected in a country like India, which has not advanced from the 
agricultural stage into the scientific and commercial, and whose ohief wealth has 
always consisted in the produce of the soil. 

75. The trading castes are enumerated in the tabular statement annexed (group 
IV). A considerable number of the castes included in previous lists are engaged in 
trade, but not in trade only ; for they are themselves the producers of the commodities 
in which they trade. Thus among the castes allied to the hunting state, there are 
many who practise the trade of butcher, poulterer, egg-seller, game-catcher, Ac, 
besides hunting and trapping on their own account Among the castes allied to the 
fishing state there are many who catch fish and make nets for sale, Ac, besides fish- 
ing for their own personal wants. Among the pastoral castes many, in fact all, trade 
in milk, butter, wool, #c. Similarly the agricultural castes retain for their own use 

Digitized by 


( 8* ) 

only a small portion of what they produoe, and dispose of the rest to men of the 
Baniya or shop-keeping classes: and there are some agriculturists, like the K&ndu and 
the Tamboli for example, who keep shops themselves for the sale of the goods raised 
by their own labour. The same remark applies to all the artisan castes, from the 
lowest basket-maker up to the jeweller, tailor, or confectioner; for many of these men 
keep regular shops, or can do so if they choose. What we mean, then, by the trad* 
ing castes, as distinct from those already enumerated and described, is that their 
special function consists in the distribution and exchange of wealth and has nothing 
to do with its production. If some of them have lately commenced to work in the 
fields (and what caste in India has not done so ?), this does not aflect the question of 
their original status as members of the Hindu community. 

76. Among the trading castes proper there is a gradation of rank from low to 
high— not so clearly marked as among the landed and the artisan castes, but not less 
certainly the result of natural laws. These gradations depend partly on the social 
status of the classes with whom they are chiefly brought into contact, partly upon tho 
degree of importance or dignity attaching to the commodities in which they trade, and 
partly on the amount of capital or wealth necessary to the practice of the trade itself. 
The very same conditions, as the reader will have observed, have determined the rank 
or social status of the several trading and commercial classes in our own country. 
The business of a green-grocer, for example, is about as low in England as it is in 
India, while that of banker is quite as high. There is no mystery in either case, and 
the three conditions which have been assigned as determining their rank apply 
equally to both. 

77. The trading castes may be arranged as follows in the order of social res* 
peciability :-*- 





, Banjara 


j Bilwar 
I Bhurtia 
r Kasondhan 





Oth ... 
I ^Maheshwari 
| Agrahari 

I Bohra 


'Pedlars and small retail dealers who 
f seldom keep regular shops. 

Traders who keep regular shops and 
f. deal in larger quantities of grain, ■ pi- 
ces, perfumes, cloth, &o., &c. 

Bankers, money-lenders, wholesale deal* 
ere, &c. 

All of the above castes, excepting the three first named (Banjara, Kunjra, and Bhnnja, 
who stand lower in the social scale than the rest) are known by the generic name of 
Baniya, which means shop-keeper or merchant The last five, who as a rule stand 
higher than the rest and trade in money rather than in commodities, are not uncom* 
monly known by the more dignified title of If ahijan or banker. AH bnt the first 
three wear the janeo or sacred thread and disallow the remarriage of widows. 

78. The above list, however, does not profess to be exhaustive. It gives only 
the main castes belonging to the general head of trader. There are many minor ones, 
whose names and characteristics it is difficult to discover, and I know of no branch 
of the subject of Indian castes which presents so many and such petty complications 
as that connected with trade. This has arisen from more causes than one. In the 
first place (with the exception of the green-grocer and the grain-parcher), there is no 
speciality in the kind of trade which any one caste is accustomed to practise, to tho 


Digitized by 


( 34 ) 

exclusion of other kinds. In this respect the trading castes of India differ essentially 
from those which, nnder the name of trade-guilds, grew np in the great commercial 
centres in Europe during the middle ages. Any trading caste in this country may 
deal in anj kind of commodity that it prefers, and hence within the same caste every 
variety of trade will be found to exist. Thus the element of speciality which in the 
case of all the Indian castes previously named has been the mainspring of their exist- 
ence, and which in England gave rise to the system of guilds or commercial castes 
known as grocer, draper, fishmonger, silk-mercer, goldsmith, &c.,is here entirely want- 
ing. In the second place, no restriction has ever been imposed by the laws or customs 
of the Hindus through which a man belonging to any of the landed or artisan castes 
could be debarred from setting up as a trader, if it pleased him. The business of 
trader has always been open to all comers alike; and no charters were ever issued 
conferring certain monopolies of trade upon certain families or clans, as was formerly 
the case in Europe, Consequently there has been a continual influx of families from 
the various industrial castes, who have detached themselves from the ancestral caste 
and calling, but have not coalesced with each other, so as to form a compact trade- 
union of their own, united by the bond of common sentiments and traditions. Thus 
the Bhunja is an offshoot from the fisherman caste, the Lohia from the iron-smith, tho 
Kasondhan from the brazier, the Pathel from the agriculturist (Kurmi), the Unaya 
from the writer (Kayasth), the Bohrafrom the Brahman, the Khattri from the warrior. 
But these offshoots have neither amalgamated with each other nor made common 
cause with trading castes already existing. The caste prejudice which they brought 
with them has survived the separation from the parent stem, and jealousy, rather than 
a desire for union, has been the feeling uppermost in their minds. Thirdly, trade is 
not a pursuit in which hereditary skill is necessary to success, whereas in the various 
arts and handicrafts described under the heading of artisan, and in many of the func- 
tions connected with hunting, fishing, cattle-grazing, and agriculture, a man has little 
chance of becoming a proficient, unless he commences to acquire from early boyhood, 
and under his father's tuition and guidance, that degree of manual dexterity and 
quickness which can seldom or never be acquired by an adult beginner. The conse- 
quence has been that while in the case of artisans, &o., there is a system of clearly- 
defined castes, each distinguished from the other by some hereditary peculiarity of 
craft, in the case of traders almost every distinction of caste that can be said to exist 
is a distinction without a difference. 

79. Notwithstanding all this, the majority of the trading castes have adopted 
names which designate in some form or other the trading function ; and thus the 
general law announced throughout this paper, that function, and not blood or creed, 
is the basis on which Indian castes have been formed, receives in the present case a 
stronger exemplification than in the case of castes previously described. Banjara, 
the name of the forest trader, is derived from banij, " merchant ;" or it might be 
a variant of banclidra, " one whose way is in the forest " : and this is the best descrip- 
tion of the function peculiar to this caste. Eunjra, the green-grocer, is from kunj 9 
a fruit-garden or vegetable garden. Bhunja, the grain-parcher, is from the root bhun 
to roast or parch. Raunia, more fully written " Bavaniya," is from ravan, which 
means " the crying or hawking of wares for sale." Kuta, the seller of husked 
rice, is from kutna, to beat out. Bilwar is probably a contraction of bela-war, being 
derived from bela> a money-bag. Bhurtia, the retail dealer and petty usurer, is from 
bharti-karna, the lending of small sums for short periods. Lohia is derived from loka 
iron, for which men of this caste barter grain and other wares. Kasondhan or Easar- 
bani (names of now distinct but once identical castes) are derived respectively from 
kansa y bell-metal, and dhan, wealth, or bani 9 seller. Orh or Rora is probably from ol 
" one who stands security for a loan." Agrahari and Agarw&la are probably from 
agari or agar, a kind of fragrant wood which is sold as a perfume. Bohra is a con- 
traction of beoJuira, "one engaged in beohar or busipass." Khattri is a variant, or 
(to speak more correctly), the modern Hindi pronMclatWpij of Kshatriya, an offshoot 

Digitized by 


( 35 ) 

of the warrior caste which took to the pursuit of money-making in preference to that 
of fighting. Among the remaining names two have a religions sense — Yishnoi and 
Maheshwari. The remaining names are perhaps derived from places. 

80. The Banjara. is the least civilized of the trading castes, and I have therefore 
placed him at the bottom of the list. As a trader he stands in the same relation to 
other trading castes that the hunter does to the other landed castes. His speciality 
consists in conveying merchandize on the backs of bullocks through trackless forest 
paths where any other class of trader would be lost. To this industry he has added 
two more— the pursuit of cattle-grazing, which connects him with the nomad or 
pastoral state, and the pursuit of robbery and rapine, which connects him with the 
hunting or savage state. The tribe professes to be descended from an illustrious 
marauder whom they call Mithu Bukhia. But what has really kept them together 
and made them look like a separate aboriginal tribe is not community of origin, but 
unity of function and character ; for it is well known that any man (or woman), of 
whatever tribe or caste, can become a Banjara and take his place among their ranks, 
provided he is willing to comply with the conditions of admission. Banjaras do not 
come within the regular line of Hindu castes ; for they have a caste or clan system 
of their own, based upon the mythical fourfold division into Brahman, Kshatriya, 
Vaisya, and Sudra. They are chiefly to be found in the Tar&i or sub-Him&layan 
districts of Hindustan and in the forests of Central India. It is not at all likely that 
tbeir clan system will last much longer ; for as a predatory or criminal class they are 
being rapidly put down under British rule, and as a commercial and carrying class 
their usefulness is slowly, but surely, coming to an end through the extension of railways 
and the opening out of roads in what remains of the primeval forest The probability 
is that the tribe will before long bifurcate, one section devoting itself to the business 
of milkman aud cattle-grazer, like the Gaddi or Gujar, and the other to that of pedlar 
or shop-keeper, like the Baunia, &c. The cattle-grazing portion of the tribe will no 
doubt take to agriculture in course of time, as a great number of Ahirs and Jits 
have already done. Already, in fact, there is a class of Banjaras called Nayik who 
have settled down to agriculture in the Benares and Allahabad districts and are 
rapidly losing the manners and traditions of the tribe. Such is the process by 
which agricultural communities are developed from cattle-grazers, and cattle-grazers 
from wandering savages and marauders. Such has been the method of caste-forma- 
tion in the plains of India. 

81. The Eunjra or green-grocer has passed beyond the wandering and predatory 
stage peculiar to the Banjara, though his status is still very low. He goes from door 
to door selling fruit and vegetables, which he carries on his head in a basket ; for he 
rarely keeps a shop. He raises no crops himself, but retails produce raised by men 
of the gardening and fruit-growing castes ; and not unfrequently he buys a crop on 
the ground before it is ripe, running the risk of loss should the crop turn out a bad 
one. The money required for such purchases is almost invariably borrowed from some 
petty local usurer, and in consequence of the risk incurred a very high rate of interest 
is exacted. Another way in which a Kunjra attempts to eke out a livelihood is by 
gathering for sale and for his own consumption the wild fruits of the earth, such as 
the plum, the fig, the mango, the custard-mpple, and many more. This indicates, what 
is the fact, that he is not far removed above the status of those wandering and semi- 
savage tribes whose natural home is the forest and to whom he is closely related in 
blood. The caste is now almost exclusively Muhammadan, but, as the name which it still 
bears implies, it was originally Hindu. This Hindu original was made up of families 
or clans taken from low tribes, such as Pasis, Arakhs, Khatiks, Ac. (vide group II.A) 9 
who still live partly by the collection of wild fruits and vegetables and still adopt 
occasionally the trade of green-grocer, but without detaching themselves from the 
tribes or castes to which they belong. Men of the Kanjar tribe, which ranks much 
lower even than Pasis, Ac, adopt the trade of green-grocer whenever they become 
Muhammadau and assume the title of Mewafarosh, the Persian name for fruit-seller. 

Digitized by 


( 36 ) 

The name Mewafarosh is sometimes used as a synonym for Kunjra. It is not difficulty 
then, to see from what class of men the green-grocer caste has sprang, and from which 
it is being even now recruited. The trade is coasidered so low that no Hinda or 
Muhammadan. of any respectable caste will enter it. 

82. The caste of Bhunja or grain-parcher is much more respectable, though still 
low as compared with the castes that follow. The Bhunja stands on about the same 
level as the Kahfir or waterman, of whom he is a near offshoot. Kah&rs still gather the 
wild rice which grows on the banks of swamps and rivers, and in my account of the 
agricultural caste called K&ndu I gave reasons for supposing that the first rice-grow- 
ers in India were men of the fishing or watermen tribes. This supposition is con- 
firmed by the fact that the Bhunja, the Kdndu, and the Kahar have many traditions in 
common, and that the three castes are sometimes confounded, because the K&ndu and 
the Kabir sometimes engage in the same trade as that of the Bhunja. There are two 
forms in which parched grain is sold — one in the whole state, and the other in. the form 
of a powder or flour called sattu, which consists of parched gram mixed with parched 
rice or parched barley. The Bhunja never mixes this flour with water, but invariably 
sells it in a dry state. The buyer or consumer must add water drawn by his own lota 
or brass pitcher, and even then the rules of Indian caste do not allow him to eat it 
anywhere but on a chauka or prepared cooking-floor. The art of the Bhunja as com- 
pared with that of the Halwai or confectioner is so extremely simple that I have 
classed the former as a trader rather than an artisan. For there is some art and 
some trade in almost every form of industry, and castes must be classified according 
to that element which predominates. 

83! The five castes named immediately below the Bhunja in para. 77 are for the 
most part pedlars or small retail dealers, who as a rule do not keep regular shops. 
Their status, therefore, ranks generally below that of the castes which follow. But 
there is nothing to prevent them from rising to the position of the highest merchant 
or banker, if their business is sufficiently prosperous. 

The Raunia (which means literally a crier) moves in a circle of some eight or 
ten villages surrounding his own ; and if he cannot get cash for his grain, he barters it 
for spices, tobacco, sugar, condensed treacle, Ac. In these days he is seldom able to 
live by trade alone, and in the intervals of business he raises his own crops. But 
trade, and trade only, was his original function. 

The Euta or Kutdmali is one who nnhusks rice and takes it about for sale in 
baskets or on the backs of bullocks. He also cultivates at times, like the Raunia and 
for the same reason. 

The Bilwar is one who acts as the weigher of any kind of market produce that is 
not sold by measure, and this has led to his becoming himself a salesman. He 
frequently takes work as a dalldl or broker ; and if the buyer and seller cannot come 
to terms, he mediates between them and sometimes advances the rupee or pioe which 
seals the bargain. 

The Bhartiya is a pedlar who deals chiefly in spices and grain and lends small 
sums on usury. 

The Lohiya is ohiefly known as one who barters tobacco, salt, parched grain, or 
uncooked grain for discarded iron, discarded clothes, old cotton, waste paper, discarded 
books, Ac. The articles thus taken in exchange are retailed for cash to any who can 
make use of them. 

84. There is nothing particular to add respecting the seven castes whose names 
stand next in the list given in para. 77. Some of them are petty dealers like the 
preceding ; others are men of wealth like the castes which follow ; but the majority 
are men of moderate means, who keep regular shops for the sale of cloth, silk, spices, 
scents, piokles, salt, sugar, grain of all kinds, Ac. As was stated above, any man of 
whatever caste may deal in any kind of commodity that he prefers. It is true that* 
many of the specific kinds of trade have received specific names, but these names 

^itized by 


( 87 ) 

must not be taken to signify that there are castes corresponding to them. Thus a 
spice-seller is called Pasari ; a perfume-seller, Gandhi ; a druggist or medicine-Seller, 
Attar ; a cloth-seller, Bazaz ; a seller of groceries, such as salt, sugar, ghi, flour, is 
called Parchun ; a seller of stationery and other small wares, such as needles, pins, 
knives, reels of cotton, &c., is called Bisati. If each of these specialities had given rise 
to a corresponding caste, the subject of the trading castes would have been much 
simpler than we find it, and this would have been more in keeping with the principle 
on which Indian castes have been generally founded. The mistake has not unfre- 
quently been made of supposing that the names of specific trades represent correspond- 
ing castes. - For instance, in the census report now under review " Gandhi " or 
"perfume-seller" has been entered among the list of castes ; and the same mistake 
was made by the late Mr. Sherring, who counted not only Gandhi, but Bisati as 
separate castes in his chapter on traders (see Yol. I, Tribes and Castes, page 302). 
In point of fact, any member of the twenty castes named in para. 77 except the three 
first may set up a scent shop and thus become a Gandhi if he pleases ; and not only he, 
but Brahmans and men of many other castes may do the same. There are two castes, 
viz., Kasondhan and Easarbani, whose original speciality (if we are to trust the 
etymology of the names) was the sale of brass utensils not made by themselves. This 
speciality still clings to them to some extent, but it has not prevented them from 
dealing in any other commodities that they please ; and there are many members of 
these castes in whose shops nothing in the shape of a brass vessel would ever be 

85. The five castes named last in this list are, generally speaking, the wealthiest, 
and hence in social status they generally rank the highest. It is from these castes 
that most of the native bankers are drawn ; and in all civilized countries capitalists 
and financiers are among the most influential members of the community. Comte, in 
the picture of an ideal state which he drew in his Positive Politique, plaoed bankers 
at the head of what he called the secular power, on account of the widespread influ- 
ence which such men exercise in the direction of industry. Certainly in India the 
Ehattri ranks almost as high as the highest Brahman and considerably higher than 
many who have assumed the Brahmanical title. 

86. The Agrahari and Agarw&la must have been originally one— sections of one 
and the same caste, which quarrelled on some trifling qesstion connected with cooking 
or eating and have remained separate ever since. I have derived the name from 
agar or agari, a kind of scent. But the etymology usually given is either " Agra," 
the city, or " Agroha," a small town on the borders of Hariana. Another etymology 
is that they are sprung from an ancestor called Agar Sen. The Agarw&a is as a 
rule a wealthy and prosperous caste. Many are bankers and usurere. Some keep 
large grain shops, others deal largely in gold and silver jewellery and keep Son6ra 
in their employment for this purpose. 

87. The Bohra seldom keeps a shop, and is almost universally known as an 
usurer, in which character he is celebrated for his unscrupulous and relentless rapacity. 
He is by origin a Brahman. But as usury is demed to be irreconcileable with priestly 
pretensions, he has been forced to detach himself from the parent stem, and to found 
a new caste. It is said that to prove his Brahmanical descent and yet to show his 
contempt for the caste which he haa forsaken, he associates with Slahd-Brahmans or 
funeral priests— a class of men whom all other Hindus avoid as unclean. It is said, 
too, that in the eastern presidency some Bohras have become Muhammadan. 

88. The Ehattri is the highest and most important of all the trading castes in 
India. The name is merely the modern pronunciation of the ancient " Ksbatriya," 
and hence the origin of the caste should not be " an ethnological puzzle/' as it is 
termed by the late Mr. Sherring. Every tradition connects them with the great 
warrior and ruling caste ; and as men of the ruling caste must necessarily be in the 
way of accumulating more wealth than their subjects, it is not surprising that certain 


Digitized by 


( 38 ) 

families should have abandoned the military life and formed a fresh caste of their 
own devoted exclusively to commercial pursuits. Khattris are much more strict in 
the observance of Hindu rites than their warlike and land -holding cousins, the 
Chattris. In this respect they are superior to many men who profess to be Brah- 
mans. A Brahman of the Saraswat tribe will eat food cooked by a Khattri, but not 
by a Chattri ; and there are hordes of Brahmans with whom a Khattri would disdain 
to associate. Khattris commence the study of the Vedas or other religious books at 
the orthodox age of eight, whereas many Chattris and Brahmans never commence at all. 
In the Panjab the priests ox gurus of the Sikhs (if the Sikhs can be said to have any 
priesthood at all) are chiefly Khattris. The Khattri is almost the only Indian trader 
who is known outside his own country. The greater part of' the trade of Afghanistan 
is in his hands, and he was seen by Yambery in Central Asia, throwing offerings on 
the eternal flame, which burns, self-kindled and self-fed, at Baku. 

89. There are two more circumstances connected with the trading castes which 
deserve a passing notice. ' One is that they are the only section in the whole of the 
Indian community in which Jainism, once the rival creed to Hinduism, has held its 
ground. In every other caste or system of castes Jainism has entirely perished, and 
yet historians are agreed that most of the men who were converted to Buddhism or 
-Jainism in its early days were from the lower castes or from tribes who had no caste. 
I can only ascribe this phenomenon to the fact that the sedentary life of the banker 
and merchant gives more leisure for the study of religion, this being almost the only 
subject to which the leisured classes in India have paid any serious attention. Hence 
while the industrial and less reflective classes were entirely won back to Brahmanisni 
several centuries ago, the commercial and studious classes are to this day disputing 
among themselves the claims of the rival creeds, as their ancestors were doibg some 
two thousand years before them. 

90. The other circumstance to be noticed is that some of the highest of these 
commercial castes, especially the Khattri, the Agarw&l, and the Dhusar, and the Jain 
caste of Oswal, have a much stronger dash of Aryan blood than any other castes in 
Upper India. If there is any marked exception to the theory announoed in para. 10 of 
" the unity of the Indian race," it is to be found in these castes. The Khattri pilgrim 
whom Vambery met at Baku is called by him " the yellow-faced Hindu." Vambery 
would certainly not have used such an expression of the average Brahman or the 
average Chattri, and yet these are the two castes which have been described by most 
writers as pre-eminently Aryan, which shows on what very loose grounds the Aryan 
theory has been held. At the 8anskrit school in Ayodbya, which is attended exclu- 
sively by Brahmans, there are representatives of this caste from many different pacts 
of India — Lucknow, Gonda, Kheri,Unao, Agra, Allahabad, Mirzapur, Chupra, Tirhoot, 
Darbhanga, Muzaffarpur, Gorakhpur, Patna, Mainpuri, Thaneswar, Malwa, Guzerat, 
Orissa, Gaya, and Travancore : and yet almost every face is as dark as that of the 
average Hindu, to whatever caste he may belong, nor is there anything at all re- 
fined or Aryan-looking in the features. On the other hand there is a " Khattri 
Patshala " in Lucknow which is attended almost exclusively by boys of the banker 
castes— Khattris, Agarw&ls, Oswals, Ac. — and here almost every face is comparatively 
fair, while the features are as refined as those of the Parsi or Cash mi ri, both of whom 
are undoubtedly descended, with little or no admixture of foreign blood, from the 
ancient Aryan. It seems, then, that the moneyed and commercial castes of the present 
day are descended from old Kshatriya families, who were much less ready to admit 
outsiders within their ranks than the priestly and warrior clans, who thrive by ex- 
pansion rather than by exclusiveness. An aristocratic moneyed class possessing no 
political power like the Chattri and no priestly power like the Brahman, and having 
nothing but personal wealth to rely upon for the maintenance of thejr social dignity 
and importance, would be more than usually careful to preserve the \purity of their 
blood. But whatever the correct explanation may be, I do not consider that the 
exceptionally fair complexion of a few of the trading castes clashes witli the general 
hypothesis on which the present theory of caste has been based, viz., the t unity of the 

Digitized by 


•< 3 M 

Indian raoe. As I stated in para. 10, traces of an admixture of alien blood mast be 
expected in isolated cases, just as the bine eye of the fair-skinned Lombard peeps out 
at times among the swarthy inhabitants of Italy. Individual ' instances v of Aryan 
atavism are not uncommon even among the lowest of Indian castes, such as the Cham&r 
or the Bhangi, who for thousands of generations past have been exposed to the scorch 
ing effects of a tropical sun ; nor is the dark skin, though comparatively rare, a thing ; 
unknown in the Ehattri caste, which has for an equally long period lived under the 
shelter of strongly-roofed houses. The main contention urged in this paper remains 
unshaken, that the Indian race is practically one in blood, character, traditions, and 
sympathies, and that caste is not a question of blood, but of function. 

91. It remains to show how far the names given in my own table (group IV) 
differ from those which were entered in the census list of castes : — 

(a) Euta and Eotamali, against which 207 and 3,232 persons have been record- 
ed, are, I have reason to believe, different names of one and the same caste ; I have 
therefore combined the figures under the name of Kuta. 

(b) Ronia, Bamaiya, and Rawa, against which 38,105, 3,369, and #3 persons 
have been recorded, are, I believe, one and the same caste, and I have included them 
all together under the name of Baunia. The root of these names is ravan, which 
means " using the voice." Hence ravaniya means a trader who cries his wares ; 
and in a previous census this caste was actually entered under the name " Ravaniya." 
By substituting u for o the name becomes Bonia or Baunia ; by another slight 
change it becomes Bamaiya ; and by a contraction of the last syllable it becomes 
Bawa. It is impossible in any case that there can be a caste of only 33 persons, 4 

ie) Gandhi, against which 66 persons have been recorded, is not a caste, aq I have 
explained already in para. 84. 

(d) Rehti, against which only 289 persons have been recorded, cannot be a sepa- 
rate caste, and I learn that it is a sub-class of Khattris, amongst whom I have there- 
fore included them. 

(e) Baniya has 1,204,130 persons recorded against it. But Baniya is no. more a 
caste than Uhirimar or fowler, Karigar or skilled workman, Kishan or ploughman, Ac. 
It is simply a generic name signifying " trader " and is applied to any man, of any 
caste whatever, who is engaged in trade, even to a Brahman or a Chattri. 

(f) No mention has been made in the census of the castes of Vishnoi, Bastogi, 
Unaya, Maheshwari, Dhusar, Agrahari, Agarwal. All of these appear to have been 
thrown in under the casteless name of Baniya. 

(g) Kolapuri is, I learn, the same as Kasarwani. The name is evidently derived 
from the town of Kolapur. 

92. T/ie serving castes. — We now come to the series of castes to which I have 
given the name of serving and which are enumerated in group V. Their speciality 
consists in ministering to the wants of man, bodily and mental ; and their rank in 
social scale depends upon the nature of the service rendered. From the list given 
below — in which, as in previous lists, the names of are mentioned in the order 
of dignity, beginning with the lowest — it will be seen that the ranks of these castes 
coincide very closely with those of the corresponding classes in our own country, 
so far as such classes can be said to exist : — 

L— Bhangi ,.. ... ... ... ... Scavenger. 

II. — Dbobi ... ... ... ... „• Washerman.. 

III.— Kahar ... ... -, ... „. Water-carrier and house servant. 

IV.— Napit or Nai Mt ... ... ,„ Barber and surgeon. 

V. — Pawariya, Dhari, Dom Mirasi, and Kathak ... Singer and musician. 

VI.— Bhst ... ,.♦ •„ ... ... Bard and genealogist,. 

VII.— Kiyasth ... ... (M ... Estate manager and writer. 

The broad line of distinction among these castes turns upon the question whether 
.they are literate or illiterate. Roughly speaking, the first four are illiterate and 

Digitized by 


' ( 40 ) 

minister to the bodily wants of men; the three last are literate and minister to their 
mental wants. 

93. Each of the above castes represents a clearly-defined social unit, and the 
names by which they are designated are descriptive of the functions which have 
called them into existence. Bhangi, the name of the scavenger or sweeper caste, is 
derived from bhang, in allusion to the imparity caused by the sweeper's touch or 
presence : Khakrob, the Persian name of this caste, signifies " dirt-brusher." Dhobi, 
the washerman, is from dhona, to wash or make clean. Kah&r, the water-carrier, is 
from two words, ka or kam, water, and Atfr, carrier : hence another, but less common, 
name for the caste is Kambar. Napit, the barber, is an abridgement from sndpita, 
.which signifies "one who causes you to take a bath," in allusion to the invariable 
custom which compels a Hindu to bathe immediately after the barber has been plying 
his art. P&waria, the musician and singer, is derived from pdnward, the cloth or 
carpeting on which men of this caste sing and play. Dhari, the name of a similar 
caste, is derived from dhdmd y to raise the voice to a high pitch. Kathak, the name of 
another musician caste, is derived from kath, to use the voice, or kathd, a dramatic 
story. Bhat, the bard and genealogist, is from bhatta, a learned man, a man of re- 
search. ' Kayasth, the writer caste, maans literally a personal attendant — that is, 
one who acts as private secretary, amanuensis, or confidential accountant to bis 
employer. This caste has acquired a preferential claim to the exercise of certain 
other literary functions which are not altogether implied in its name. 

94. The lowest of all the serving castes is the Bhangi or sweeper. He is a 
near offshoot from the savage Dom or Ghand&l, whom in group I. I have placed among 
the casteleds tribes, and whom Hanu fitly described, some two thousand years ago, in 
the following words : — " Their abode must be out of the town ; they must not have the 
use of entire vessels ; their sole wealth must be dogs and asses ; * * * * continu- 
ally must they roam from place to place" (Chapter 2L, verse 52). To this day the 
Bhangi, faithful to the traditions of his origin, takes charge of his master's dogs, 
and like them eats what he can get from the leavings of his master's table. The 
function which has specially called him into existence as an Indian caste, and 
detached him as such from the ancestral tribe, is that of sweeping houses and streets 
and removing everything that is dirty or unclean. He is himself, therefore, the 
type of all uncleannes? ; and to persons of the higher castes his touch, even his 
presence, is considered a pollution. This is why he has received the name of Bhangi: 
for bhang means literally interruption, breaking, damaging ; and whatever a man is 
doing, if he is touched by a sweeper, he must at once leave off doing it and go and 
bathe. It is thus a common custom amongst Hindus, if they wish to remove an 
intruder or to annoy a neighbour, or to compel some one to listen to their demands, 
to send the sweeper at him, especially at such times when it is least convenient to 
him bo be disturbed. Many Bhangis have turned Muhammadans, and in the Panjab 
some have become Sikhs ; but no respectable Muhammadans or Sikhs will allow him 
to eat with them, although both of these creeds distinctly avow the equality of all 
men. His change of religion has, therefore* profited him nothing. He is still the 
Bhangi, or " one whose impurity disturbs you." I used to think that the name 
Bhangi was derived from the intoxicating drug called bhanga, and known to Euro- 
peans as Indian hemp. But as the sweeper caste is not more addicted to the use of 
this noxious plant than any other, the previous explanation is more likely to be right. 
There is another name of the caste, Chuhra, known chiefly in villages and in 
districts west of the Ganges, which probably means "vermin-catcher." 

As the broom does not always afford sufficient scope for his energies or for his 
maintenance, he practises many of the callings inherited from his near ancestor, the 
savage Dom, with whom, however, he now disdains to associate. He makes baskets 
and brooms for sale and for his own use, digs vermin out of the earth for food, 
draws sinews out of carcasses for fixing the weaver's brush or the peasant's winnow- 
ing- pan, takes off the hides of dead kids and. goats, manufactures drums with their 

Digitized by 


{ 41 ) 

skins, and plays on the rustic flute called shahndi. By the lower castes he is some* 
times hired out as a musician to enliven the celebration of a marriage ceremony* 
On occasions of public festivity he mounts a temporary platform made of bamboos 
and beats the drum and blows the flute for the delectation of the people and for bis 
own fee. He is notoriously addicted to thieving and drunkenness and to the violent 
treatment of his own women. If other means of living fail, bis daughters are sent 
out to earn money as prostitutes. He will accept of alms on days of eclipse, when 
the demon of darkness is abroad. He is supposed to have some secret understanding 
with Sitala, the goddess of small-pox; and to many classes of the community, Muham- 
madan as well as Hindu, he is the recognized hereditary priest of that abhorred god- 
dess. Such is the typical Bhangi of the Hindu village. But some families of this caste 
have in recent years acquired much higher tastes and notions. Those who have served 
in Europeans 9 houses (for it is here chiefly that the higher tastes have been formed) 
are gradually developing into a separate caste. Such persons will only intermarry 
among themselves and decline to associate with their rougher caste-fellows. Some of 
the women have become lady's maids and have accompanied their mistresses to Eng- 
land, where they pass muster with English maid-servants. 

The patron saint of Bhangis is Lai Guru (called by Muhammadans Lai Beg), 
the prince of scavengers. Hence Hindu sweepers are called Lai Gurus, and Muham- 
madan ones Lai Begis. Another Muhammadan name is Mehtar, which means 
literally u prince," a title of respect conferred upon the head of a clan, and hence 
transferred to the caste in general. Both Hindu and Muhammadan sweepers 
worship their patron saint in the same kind of way and on the same occasions* 
His presence is symbolized by a pole and flag. Swine are sacrified in his honor to 
the strains of the flute and the loud beating of the drum. 

95. The Dhobi, or washerman, represents an impure caste, but one many 
degrees higher than that of the Bhangi, from whom he has sprung. Both are des- 
cended from the Dom, " whose sole wealth,** according to Mann, " must be dogs and 
asses." The Indian washerman has always been associated with the indigenous 
ass, which carries the soiled clothes down to the bank of the river or tank and takes 
them back clean to the house. No Hindu of any caste, even the lowest, will wash 
his own clothes ; and so the Dhobi has been formed into a caste which shall bear 
the impurities of all. He is not admitted into the inside of a Hindu temple, and 
this at once places him on the same social level as that of the " castes allied to the 
hunting state " and the lowest castes of artisans. His work, however, brings him 
into continual contact with one of the purest of elements, and the quiet nature of his 
occupation has taught him a certain refinement of manner. For these reasons he 
is far less despised than his Bhangi cousin. He has left off eating pigs, as the 
Bhangi himself has left off eating dogs, while their common ancestor, the Dom or 
Chand&l, still eats both. The washerman is as necessary an element in the constitu- 
tion of the Hindu township as the watchman, the hide-tanner, the blacksmith, the 
potter, and the carpenter already described, and he is remunerated, as they are, by 
a portion of the produce of each half-yearly harvest. On days of child-birth, and 
on the day when the mother first leaves her room, he receives a special fee in cash or 
kind t for the garments and wrappings worn till then are believed to be something . 
more than physically impure, and no one but a Dhobi can make them fit for fixture 
use. Many families of this caste have become Muhammadan, but this has not 
materially improved their social status; nor, in spite of the prohibitions of their 

- adopted creed, has it checked their propensity to drunkenness. 

96. The Kah&r is the general house-servant in respectable Hindu families. His 
primafy function is that of drawing and bringing water for the bath or the table, 
and this has led to his being employed for various other uses, such as taking care of 
clothes, dusting the rooms, kneading the ohap&ti preparatory to cooking it, carrying 
the palanquin, &c. The two castes previously named—the Bhangi and the Dhobi — are 
sprung from hunting tribes of the stamp of Dom or Chand&l. The Kahdr, on the 


Digitized by 


( 42 ) 

other hand, is sprang from fishing tribes, such as the Gaud, the Dhuria, the Dhimaf, 
and others mentioned in group II. B. As was stated in para. 22 (e), it is not unusual 
to speak loosely of all the fishing tribes as Kah&rs, because any man from among 
these tribes may take employment as a water-bearer, if he can find a master. 
Properly speaking, however, the name Kah&r belongs only to those families which 
seceded long ago from their ancestral tribes, abandoned the hereditary industry of 
fishing, and formed a new and distinct caste devoted exclusively to domestic service. 
The best evidence of the existence of such a caste is that it has a Muhammadan 
counterpart —the Bhisti, whose functions in a Muhammadan household are pre- 
cisely similar to those of a Kah&r in a Ilindu establishment. The only difference 
between them is that the Kah&r invariably carries water on his shoulders in a pitcher, 
while the Bhisti carries it across his back in a leather bag, called a masak, which is 
made of the hide of a bullock or large goat. Orthodox Hindus will not even bathe 
in such water, much less drink it; for they consider the contact of leather a pollution 
to such a pure element as water. In European houses the Kah&r (who is called 
bearer, a contraction for water-bearer) is used for general house-work, and the 
Bhisti for drawing water. In Southern India the name for the corresponding caste 
is Bhoi, which has been corrupted by Europeans into the Madrassi "boy." Both 
the Kah&r and the Bhisti are respectable and orderly castes and are seldom or never 
seen inside a jail. The worst legacy that the former has inherited from the ancestral 
fisherman is his propensity to drunkenness. A man convicted of directly stealing 
his master's property, (for indirect gains at his master's expense are not condemned), 
is expelled from his caste, and can only be re-admitted on payment of the fine or 
giving the banquet decreed by the caste assembly. 

97. The N&pit, or hair-cutter, is, like the Uhobi, an essential element in the 
constitution of the Hindu township. No village community could exist without him. 
All Hindu castes who are respectable enough to be admitted inside a temple (and 
even castes lower than these, if they can pay the fee) employ him for certain ceremo- 
nial observances connected with infancy, burials, and marriages, and most of these 
ceremonies are enforced by religious sanctions. 

Every child after the age of six months or a year undergoes the ceremony of 
having its head for the first time touched with the razor. This is a ceremony of no 
little importance in the eyes of a Hindu and is called chura karan. It is performed 
in the presence of some deity, or rather in that of his image, who is believed henceforth 
to take the child under his special patronage. The cutting off of the birth-hair is 
believed to remove the last trace of the congenital taint inherited from the maternal 
womb, and hence the ceremony has the same significance as that of a baptismal or 
lustral rite. The custom of cutting off the birth-hair has been widely practised among 
backward races elsewhere, and is not at all confined to the natives of India. The 
germs of the barber caste may, therefore, have existed in times before the Aryan had 

In funeral ceremonies the N&pit plays an important part. He shaves the head 
and pares the nails of the dead preparatary to cremation. He shaves the head of ttuo 
man who puts the first light to the pyre. Ten days afterwards, he shaves the head of 
every member of the household. By this time, after taking a final bath, they are 
purified of the contagion of death. 

In the celebration of marriage ceremonies he acts as the Brahman's assistant, 
and to the lowest castes or tribes who cannot employ a Brahman ho is himself the 
matrimonial priest. The important part he plays in marriage ceremonies has led to his 
becoming the match-maker among all the respectable castes. It is he who hunts out 
the boy, finds out whether his clan or caste is marriageable with that of the girl, 
settles the price to be paid on both sides, takes the horoscopes to the Brahman to he 
compared, so as to see if the stars are favourable, carries the presents from one houso to 
the other, and so forth. His function as match-maker is not an unimportant one ia a 

Digitized by 


( 43 ) 

state of society in which the rules of caste have imposed endless restrictions on th* 
freedom of marriage. 

Shaving is not the only service that he is expected to render to his constituents. 
He is the ear-cleaner, nail-cutter, cupper and bleeder, Ac. In short, he performs any 
kind of operation on the body of man that requites a sharp knife, from shaving a 
beard to lancing a boil. He might therefore be fitly styled a barber-surgeon. In 
this double capacity be is the exact counterpart of the barber-surgeons of mediaeval 
Europe, out of whom the modern medical profession has sprung. The Muhammadan 
Mpits (who are called Hajjam) are said to be more skilful in surgical operations 
than Hindu ones. Like Kam&ngars (see para. 63), they are largely employed in 
setting dislocated bones. 

His wife acts as nurse to the mother and child for the last six days of the confine- 
ment During the first six days they are in charge of the midwife— some woman 
of the less respected castes of Cham&r, Dh&nuk, or others. 

98. The castes of musician and singer fall naturally into two main classes— the 
Muhammadan and the Hindu ; and the list may be shown as follows :— 

Piwariya ... — 1 

Dhari ... - > Muhammadan. 

Dom Miraii ... ••• ' 

Kathak — ~ Hindu. 

The names by which the former are called, being of Hindi and not of Persian origin, 
are alone enough to show, even if other evidence were wanting, that these castes were 
originally Hindu ; and this is my reason for recognizing them as a part or product of 
the general Hindu system, as I have done in the parallel cases of Eunjra, Kam&ngar, 
and Tirgar, all of which are now Muhammadan. In fact there are but few Hindu 
castes to which there is no Muhammadan counterpart ; and in some cases this counter- 
part has become the larger element of the two, while in others it has ousted the Hindu 
element altogether. It would thus be impossible to give a complete outline of Hindu 
castes without including those in which the Hindu element has now either partially or 
wholly disappeared. 

The Pawariya caste has one speoial function. On the twelfth day after the birth of 
a child, when the mother receives her final bath before leaving the lying-in chamber 
and puts on a new wedding bracelet, and when all the pottery ware of the household is 
broken up and replaced by a new set, the P&wariya comes uninvited to the house with two 
or three comrades of his own class, female as well as male, and sings songs of gratu- 
lation to the strains of a fiddle. The fiddle is called the kingri, and from this instru- 
ment the caste has also been called Kingriya. The instrument is played with a bow 
and consists of strings of sinew or steel stretched over a hollowed gourd or calabash 
(tomri) as the sounding-board. In return for this unasked attention the performers 
receive presents of grain or money and clothes, which last they afterwards sell to somo 
Bauniya or other pedlar, if they have no need of them. 

The Dhari oaste (which appears in the census report under the less common name 
of Dhariwal) is less specialized than the preceding and more migratory. They go 
from place to place, wherever they can find employment, and sing and play at marriages 
and sometimes at Hindu temples. Their chief instruments are the double drum called 
mridang and the tambura, a kind of guitar which is played with the fingers. Both 
men and women perform on these instruments. 

The Dom, or as he prefers to be called the Dom Mir&si, is one who is especially 
attached to a certain circle of houses as the family musician and jester. The word 
Mirasi means "hereditary", and the caste has been so called because the son inherits his 
father's connexion. The instruments most commonly used by the men and women 
of this fraternity are the light drum (dholak), *nd brass cups or cymbals called majira. 
The men, too, sometimes play the fiddle called kingri. The women dance, but only 
before the ladies of the zenana ; and they are sometimes regularly entertained as jester* 

Digitized by 


( 4* ) 

to help these ladies to kill the time and reconcile them to their domestic prisons. Of 
each of these three castes it must be said that the women do not dance before men 
and hence they are reputed to be chaste : for no woman who is not a prostitute will 
dance in the presence of the opposite sex, though singing and playing are not equally 

No doubt need exist as to the source from which these castes have sprung, or as 
to what their character was before they became Muhanimadan. They are offshoots 
from the wandering and casteless tribes, such as Dom, Kan jar, Kinchan, Ac., whom I 
have placed in group I, and from the lowest oastes of Hindus, such as Pasi, Arakh, Ac., 
whom I have placed in group IIA. All these tribes and castes are addicted to 
singing, dancing and playing on the drum, the fiddle, the guitar, 4c. It would be 
easy to show that similar instruments were found in use among most modern savages 
elsewhere, when the first Europeans visited them ; and as I have lately mentioned in 
reference to the Bhangi or sweeper (see para. 94;, music and dancing are among 
the means by which he ekes out a livelihood at the present day. Now the Bhangi 
is certainly an offshoot from the savage Dom, and this very name — Dom —has clung 
to the Mirasi caste up to the present day. Another link in the connection is the fact 
that the P&wariya caste to this day, in the intervals of business, maintains itself by the 
manufacture of fans, palm-leaf umbrellas, baskets, &c— the very industry in which 
the Dom and kindred tribes are known to excel. Nothing could be more natural, or 
more in accordance with the method by which Indian castes have been generally 
formed, than that certain families or clans from these tribes should have detached 
themselves from the ancestral stem and formed a new combination devoted especially 
to music, singing, and dancing. 

There is even a well-attested tradition as to the means by which they became 
converted to Islam. It was in response, as the story goes, to aq invitation from 
Amir Khusru, the poet and musician, who lived in the reign of Ala-ud-din Khilji, 
A.D. 1295. Up to that day, it is said, there were no regular musician clans or 
castes who could take service in Muhammadan houses, and whose women could be 
admitted within the zenanas. The conversion of these clans to Islam has greatly 
raised them in social estimation. The Dom Mir&si has acquired a refinement 
of manner and expression which places him in marked contrast with the savage 
ancestor to whom he is directly related in blood. Not knowing what the word Mir&si 
means, he has contracted it into Mir, a synonym for Sayyad. So while the one is 
an outcaste, eating carrion and dogs, the other is claiming to be a descendant of the 
great Prophet of Mecca* 

In the census report Dom the Musician has been mixed up with Dom the 
Savage, as is clear from the description of the occupation written against the name 
Dom: "bamboo basket-making, singing, and dancing." Mr. Williams, however 
drew attention to the difference in Oudh Census Report, Vol. I, page 83, where he 
says " there appears to be no connection between these Doms (the Muhammadan 
musicians) and the low outcaste tribe of Hindus known by the same name, whose 
employment is that of making ropes and mats." He is right in drawing a marked 
distinction between these two tribes a$ they now are, but mistaken in supposing that 
they have no ancestral connection. The development of the various castes out of the 
primeval savage is, as I think, the only hypothesis on which the formation of oastes 
and their gradations in relation to each other in the scale of respectability can be 

99. The Hindu caste of musician called Kathak is entity" distinct, both in 
origin and character, from the preceding Muhani^- aan ones « While the latter 
have ascended from the Dom and became Muh?»- iima< * an > ^ e Kathak has descended V 
from the Brahman and remained a stanch "* n( lu- In the early or Vedic days of \ 
Hinduism there were four classes of prie~* or sacrifioers— the lowest or Adhvaryu 
wbo slaughtered the victim, the Ud/ tri who ch »ated the hymns, the Hotri who 

Digitized by 


< 45 ) 

threw the offering on' the flame, and the Brahman who superintended the entire 
ceremony. Bat the bloody sacrifices of the Vedic age fell into disrepute through 
the teaching of Buddha ; with the spread of Hinduism further east the Vedic 
ritual went out of use, the Vedic hymns were forgotten, and the order of- priests 
who chaunted them (the Udgatri) ceased as such to exist. The Kathak casta 
is the descendant of this extinct order. The tradition of their origin has been 
preserved in various ways. They still wear the janeo or sacred thread ; and 
in salutin'g any one they do not make a bow, as other castes do, but pronounce 
the Asirbad or blessing like Brahmans. Though they have ceased to channt the 
Sama Veda in the presence of bloodstained altars, as their ancestors did in 
the earliest times, they are still employed to chaunt sacred melodies before idols or 
other symbols in the temples. The modulations of the voice are now, as then, 
accompanied with certain gesticulations of the arm and movements of the body which 
are tantamount to dancing. They dance before the idol, as King David danced before 
the Ark as it was being led up to Jerusalem. 

But in these degenerate days some of the functions exercised by the Kathak are 
much less dignified than those associated with religion or religions worship. The 
men are hired out to play and dance and sing at marriage festivals before large and 
mixed audiences ; and their own wives occasionally, but rarely, sing in public. The 
men are generally accompanied on such occasions by women of loose reputation, who 
dance and sing before persons of the opposite sex, shaking the castanets. Such women 
are generally of the low and profligate tribe called Kanchan, whom in group I. 
I have included among the outcaste or casteless tribes. They frequently take lessons 
in dancing and singing at the Kathak's house and even fro.m the Kathak's own 
wife. This is a remarkable instance of extremes meeting. A BrahmanioaL caste like 
the Kathak has become associated with the lowest women in India, while the lowest 
savage in India — the Donx— has risen to the rank, of a Mirasr and calls himself a 
Sayyad or descendant of the Prophet 

The caste of Kathak is slowly, bat it seems surely, dying out- The casteless 
tribes, such as Kanchan, the low-caste Hindus, such as Bhangi, and the Mahammadan 
castes described in the previous paragraph, appear to provide all that the community 
requires in the way of danoing, singing and music, and they charge lower fees. 

100. The Bhat is another caste which, like the Kathak, k an offshoot from the 
ancient Brahman. This caste, however, has sprung, not from an order of priests like 
the Kathak, but from those secularized Brahmans who frequented the courts of 
princes and the camps of warriors, recited their praises in public, and kept records of 
their genealogies. Such, without much variation, is the function of the Bhat at the 
present day. The ancient epic known as the Mah&bh&rata speaks of a band of bards 
and eulogists marching in front of Yudisthira as he made his progress from the field 
of Kuru-kshetra towards Hastinapur. But these very men are spoken of in the same 
poem as Brahmans. Nothing could be more natural than that, as time went on, these 
courtier priests should have become hereditary bards, who seceded from the parent 
stem and founded a new caste bound together by mutual interests and sympathies. 

So Bhat for several centuries past has pretended to call himself a Brahman, nor 
would such pretensions be recognized anywhere in India. But there are many 
circumstances which recall the tradition of his origin. The most important of the 
subdivisions of this caste is called Birm-Bhat, which is evidently a corruption of 
Brahma. There are some Brahmans, especially of the Qaur tribe, who still exercise the 
function of bard and genealogist, though the Bhat is never permitted to exercise the 
function of priest. Among all the lower classes of the population the Bhat is still 
called Mah&raj or great king, a title which is generally conceded only to Brahmans. 
The caste still wears the janeo or sacred cord. But the strongest proof of all lies in 
the survival of certain titles connected with the celebration of sacrifices — a purely 
Brahmanical function. The prince, chieftain, or landlord, who patronizes the Bhat 
and entertains him as his family bard is called Yajaman, or " he who gives tho- 

Digitized by 


sacrifice," while the Bhat himself is called Fagwd 9 Jajak, or Jachak, or "the 
priest by whom the sacrifice is performed." Such words recall the day, now passed 
away for ever, when costly public sacrifices were given by kings and rulers as 
thanksgivings for some great deliverance or in celebration of some great public 
event, and when the praises of the giver were solemnly recited before the audience 
by the superintending priest. 

A distinction sprang up, after the caste had been formed, between the Birm- 
Bhat, who merely recited the praises of ancestors at weddings and other festive 
occasions, and the Yagwa-Bhat, who kept the family records of births, deaths, and 
marriages. The former are of more migratory habits and are hired only for the 
occasion, while the latter have an hereditary circle of constituents, to whom they pay 
periodical visits, receiving the usual fee in return. Most of the great Chattri families 
still keep their Jagwa-Bhac or genealogist, but the practice is becoming less com- 
mon every year, and the Bhat has been falling lower and lower' with the declining 
power of the Chattris, for whose sake they came into existence. In these days a 
high place allotted at a darbar held by some Viceroy or Lieutenant-Governor does 
more to ennoble a native chief and to stamp the respectability of his family than the 
pompous recitations of a Bhat at a marriage festival. Some members of the caste 
have now taken to the plough ; but they are mostly known at the present day as 
rapacious and conceited mendicants, too proud to work, but not too proud to beg. 
The period of their greatest prosperity was that in which their services were most 
needed, viz., when the great nomad and savage chieftains of mediaeval India, sprung 
as they were from the numerous indigenous tribes whose names they still bear were 
rising into the ranks of Chattris, and when men, publicly recognized as the bards and 
genealogists of the great, were required to give official testimony to the alleged purity 
of their descent from some illustrions saint or warrior who lived in the heroic age. 
It is no wonder that the Bhat is dreaded by Chattris to this day on account of the 
power he possesses to disclose a family secret or make them the subjects of public 

The caste of Bhat, like that of Eathak, appears to be gradually dying out in 
the plains of Hindustan. In Bajputana it is still numerous and flourishing. 

101. The last and at the present day the highest of the serving castes is the 
Kayasth or writer. There is no reason to doubt that this caste is chiefly an offshoot 
from Chattris. According to the legend told in the Skanda Purana they are descend* 
ed from the posthumous son of Chandra Sen, the great Kshatriya king, whose 
widow had fled for refuge into a hermitage after he himself and all other Kshatriyas 
had been slain by Parasu Kama. The life of the widow was spared on condition that 
the son to be born from her should renounce the calling of the warrior and take to 
that of the writer. • Kayasths have from time immemorial been allowed to wear the 
sacred cord, and many of them wear it still. The name of Thakur or lord, which 
is by courtesy the title of Chattris, as Pandit or Mahfiraj is that of Brahmans, is not 
uncommon among men of the Eayasth caste. Local traditions are not wanting of 
Kayasths who have won distinction as warriors and leaders of military bands. It was 
decreed in certain law books that the Eayasth appointed by the king as his accountant 
or secretary " must be one versed in the Shastras or sacred literature," which shows 
that he was not a man of the so-called Sudra or servile class, as his detractors have 
tried to prove. It is not difficult to conceive that princes and the owners of landed 
estates generally would prefer to appoint their own younger sons, or nephews, or any 
other near relatives, to whom they had no land to give, but on whose honesty they 
could rely, as their estate-managers and accountants, and that families or clans engag- 
ed in such work for several generations in succession would gradually become detach- 
ed from the parent caste and found a new one of their own. Kshatriyas and Chattris 
have always been addicted to a love of flesh diet and the use of spirituous liquors, one 
of the points in which they differ from Brahmans — and this, too, is one of the charac- 
teristics of the Kayasth caste. 

Digitized by 


( 47 ) 

The original function, then, of the Kayasth caste was that of estate-manager, 
and this function they have retained up to the present day through all changes of 
government— from the ancient Hindu raja to the various Muhammadan dynasties, 
and from these to the English rule. From the earliest known times every Hindu 
township had an accountant or estate-manager on its staff of functionaries, just as 
it had its watchman, messenger, potter, ironsmith, carpenter, washerman! barber- 
surgeon, astrologer, &c. His work consisted in keeping the rent-roll on behalf of 
landlord, exacting the rents, drawing up leases for tenants, apportioning income and 
expenditure among co-sharers of estates, Ac, and his work as patw&ri or accountant 
is much the same even at the present day. The substitution of Muhammadan for 
Hindu rule made little or no difference in his status. But under British rule his 
position has become somewhat complicated, and he scarcely knows now whether he is 
the servant of Government or of the landlord. He is in point of fact something 
of both, especially as the Government is now beginning to guarantee him a fixed 
salary. The alphabet or character which he uses in all his registers and accounts 
is a running hand, called Kaithi, derived from N&gri, the character pre-eminently used 
by Brahmans. In the North- Western Provinces her has been forbidden to use this 
character in the rent-rolls intended to be filed in the treasury offices, and Nagri or 
Urdu have been prescribed in its place. In Oudh, however, Kaithi is still permit- 
ted. In both provinces, and in fact throughout Upper India, Kaithi is in all private 
transactions more widely used than any other vernacular character both by Hindus 
and Muhammadans. 

The Kayasth has not always served in the humble capacity of village account- 
ant From the earliest times there was a higher and more ambitions class, vrho 
served as secretaries and finance ministers in the courts of kings. In the Smriti, or 
collection of laws, ascribed to Vnhaspati, it is said that "a king has ten constituent 
parts, viz., himself, the chief justice, the counsellors, the Smriti or laws, the account- 
ant the writer, the gold, the fire, the water, and the attendants." Now the chief 
justice and the counsellors were Brahmans, but the accountant and the writer 
were Kayasths. In the same code it is said : " the accountant makes calculations, 
and the secretary takes down the proceedings of trials." The same code says in 
another place : " Two persons must be appointed by the king, a secretary and an 
accountant, who are skilled in the expounding of words and meanings, adepts at 
counting, upright, and learned in the different dialects." Quotations to this effect 
might be multiplied. 

Thus from the very first there were two classes of Kayasths — one the accountant 
of the village commune or township ; and the other the scribe, finance minister, or 
accountant in the king's court After the Hindu dynasties had been succeeded by 
Muhammadan ones, Kayasths of the latter class devoted themselves to the study of 
Persian and abjured the ancestral Kaithi, while Kaithi still remained, and remains, 
the favourite character of the patwAri or village accountant. This is why Kayasths of 
a certain class are distinguished above all other Hindu castes for their profioiency in 
the Persian language. Having thus secured the patronage of their new masters, the 
Muhammadans, they gradually ousted the Brahmans from those posts which they 
had been accustomed to occupy as judges, counsellors, Ac, under the old Hindu 
dynasties, and the decline of the political power of the Brahmans can be largely 
ascribed to this fact. 

Under British rule the Kayasths of this class have shown an equal readiness to 
learn the English language and adapt themselves to the new order of things. There 
is no caste in Upper India so largely represented in the revenue offices, the courts of 
justice, and in the literary professions, such as pleader or native barrister. These 
Anglicized Kayasths are now, in their turn, recoiling against Persian, as their fore- 
fathers did against Kaithi, and are seeking to cevive the study of Hindi and Sanskrit 

Digitized by 


( 48 ) 

All Kayasths, of whatever section or class, revere the memory of a patron saint 
called Chitragupta, whose twelve sons are considered to be the founders of twelve great 
Kayasth families, as the twelve sons of Jacob founded the twelve tribes of Israel. The 
legend of Chitragupta is that he sprang oat of the body of the Supreme Being, 
Brahm&, while he (Brahma) was meditating on some means by which the good and 
bad actions of men could be registered, fie is thus the recording angel of the 
Hindu pantheon — a function suggested by the fact already alluded to, that the 
Kayasth in the court of a Hindu king was " the secretary who took down the 
proceedings of trials." 

102. Among the list of serving or professional castes it may seem strange that no 
niche exists for the physician, whose function as a preparer and administerer of medi- 
cines is quite distinct from that of the Napit or barber-surgeon, and is considered 
much more respectable. There was, however, a caste of physician called Baidya, a 
name derived from vidya or science : hence Baidya meant "a man of science. " The 
caste still exists, I am told, in Beh&r and the Lower Provinces. But it has died 
out in Hindustan, owing to the superior reputation of Muhammadan physicians called 
Hakims, which name also means "man of science." All Hindus of respectable castes, 
as well as Muhammadans, who can aftord to pay the fee, employ an Hakim, or take their 
complaints to some native doctor trained in English medicine. It is very probable 
that many of the Hakims of the present day are descended from Baidyas, who adopt- 
ed the creed of Isl&m several centuries ago, while the Muhammadans were still in 
power and their reputation for medicine at its height. 

Among the lower, poorer, and more ignorant classes of the population, whether 
Hindu or Muhammadan, a different practice exists. Such persons begin by leaving 
their illnesses to nature, and when this fails, resort to the numerons traditional remedies 
that are current among the people, or send offerings to temples, or employ an Ojha or 
Nyotya— that is, an exorcist — to expel the evil spirit. Moreover, there is a class of quacks 
calling themselves Baidyas, who wander about from village to village and sell vegeta- 
ble drugs prepared from certain trees and herbs which possess medicinal properties. 
The vagrant and casteless tribes whom I have included in group I. are specially cle- 
ver in the knowledge of the properties of herbs. It is they who supply the shops of 
the attar or druggist with many of the native medicines. There is a Banmanush now 
in the Central Jail, Lucknow, who boasts that he is a regular medical practitioner among 
a certain circle of villages near which his gang resides. The fact that these men call 
themselves Baidya shows that there was once a respected caste which they aspire to 
indtate and to which they profess to belong. 

The name by which the physician caste is called in Mann's Code is Ambastha (see 
chapter III, 152 ; IV, 212 220 ; X, 8, 13, 47). His account of their origin implies that 
they were an offshoot from Brahmans, though in his own day they were quite a distinct 
caste. There are a few learned Hindus at the present day who stu<ly the Sanskrit 
works on medicine and who are called Baidyas by profession. But these men are not 
Baidyas by caste, but Brahmans. There are four such men in Lucknow and many 
more in Benares. 

103. The differences between the names occurring alphabetically in the census 
report and those shown in my own table (group V) are the following: — 

(a) Chera is entered in the census table as a sweeper caste of nine persons !. I 
have included these amongst Bhangis. Perhaps Chera is a misprint for Chuhra 
the name commonly given to the sweeper caste in Hindu villages. 

(b No mention is made in the census report of the serving caste of KahAr or 
water-bearer, whioh, as I have explained already in para. 96, should be separated fromi 
the fishing tribes or castes, out of which the serving caste was formed. 

(c) The P&wariya caste is not mentioned in the census report ; but mention i* 
made of a. caste called Paria,. which, I presume, is intended for this. 

Digitized by 


( 49 ) 

(d) The caste of Dom Mir&si is not mentioned in the census report. The omis- 
sion has been alluded to already in para. 98. 

(e) Bddi, against which 995 persons have been recorded, is shown as a separate 
caste. But Badi is merely a synonym for Birm or Brahma- Bh&t, and means literally 
"reciter/' a correct description of the function of the Brahma-Bhat as distinot from 
that of the Jagwa-Bhat (see para. 100). Moreover, Badi is wrongly described in the 
census report as a singer and dancer. He is in fact neither, but a reciter of family 

(/) B&j Bhat, against which 249 names have been recorded, is not a separate 
caste, but, like the preceding, a section of Bh&ts. The Raj-Bh&ts and the Badis have 
all been included under the name of Bh&t in my own statement. 

fa) No mention is made of the caste of Eathak in the tables of the census report. 

140. The priestly cartes. — We come now to the group or series which closes the 
long list of Indian castes proper and which in point of social rank stands higher 
than all that have gone before— those connected with religion. These castes may 
be almost summed up in a single word— Brahman. The Brahman caste is the coping- 
stone to which all the other stones of the social arch converge, the model upon 
which all the other castes were formed, and in a certain sense (as will be shown 
hereafter) the cause of their existence. But the word " Brahman/* whatever its 
original sense may have been, is now a term of rather vague import. It denotes 
not one, but a multitude *of castes or sub-castes, which are almost as distinct from each 
other as those of any other series or group previously described ; and if they are called 
by the, generic name of Brahman, it must not be understood that they intermarry or 
eat together, as if they constituted a single caste, but merely that they have some 
connection, direct or indirect, with religion, the common element, and have sprung 
out of a common germ. The lowest of these Brahmanical castes are despised by some 
Brahmans themselves and are held in very little respect by the upper castes of the 
outside community; but by the lower and more ignorant classes almost every Brah- 
man, whatever his rank or status may be among his own brethren, is regarded with 
. a feeling of instinctive awe, such as no other caste in the Indian community can ever 
expect to inspire. 

105. According to the census report under review the total number of Brahmans 
in the North- West and Oudh (all kinds included) amounts to no less than 4,690,850, 
or about 12 per cent, of the total indigenous population of the province, Hindu and 
Muhammadan. • Distinot altogether from the Brahman, yet closely connected with 
him in character and representing tenets and modes of life sanctioned by Brahmanical 
authority and example, is the Goshayen, whose caste numbers altogether 118,308 
persons. Distinct both from the Goshayen and the Brahman, yet in some respects 
resembling both, there are many different orders of religious mendicants, who profess 
to have retired from the world, and to have done so for the sake of studying or prac- 
tising one aspect or another of the many-sided creed, of which Brahmans have been (he 
principal authors. These, according to the census, number 226,122 persons. The 
Brahman, then, is by far the most numerous as well as the most important of the religi- 
ous fraternities, and the attempt which we must now make to sketch his origin and 
functions will take up a disproportionate amount of space in this paper— an emblem of 
the caste itself, which has appropriated far too large a share of the time, wealth, and 
attention of the rest of the community. 

106. The original function which called the Brahman into existence and formed 
him into a distinct social unit was the performance of sacrifice. It was so with the 
caste of Levite among the people of Israel, and has been so with every other order of 
priesthood throughout the world founded on a deistic basis. As regards the Brahman 
or Indian priest, this is abundantly clear from the ancient Vedio hymn-books or 
sanhitas, from the Vedio liturgies or Brahmanas (all of which deal exclusively with 
sacrifice), from the ancient law-books or Smritis, from the two great national epics 
(the Bamay&na and the Mah&bh&rata), and from many of the Pur&ns or ancient 


Digitized by 


( 50 ) 

histories. The name " Br&hman" shows clearly enough that the origin of the caste, 
so termed was sacerdotal — that is, functional, and not tribal. One of the numerous 
names by which a sacrificial prayer was called in the Vedas is brahma (neuter), 
and a man who composed or recited such prayers was called brahma (masculine) 
At the outset any man, to whatever tribe be belonged, or whatever his occupation 
might be, could be his own priest, and no such thing as a distinot caste possessing 
the exclusive ny>u,opoly of performing sacrifices existed* The original ceremonial 
practised by the first Aryan settlers in India was very simple and intelligible a* 
compared with what it afterwards became in an age less devoted to the attractions 
of war and more prone to listen to the gloomy suggestions of priestcraft. The 
divinities worshipped by the earliest Hindus were the various elements, powers, or 
departments of nature, whom they personified into a class of divine spirits or deities 
and invoked under such names as Indra, the sky ; V&y u, the wind ; Agni, fire ; Rudra, the 
tempest ; Vishnu, the sun ; Prithivi, the earth ; Chandra, the moon; Ushas, the dawn*; 
Varuna, the water, &c, Ac. As the worshipper was brought face to face with the elements 
and forces of the world around him, and was not bound to distinguish between the 
physical agents themselves and the unseen spirits or deities who were supposed to 
preside over them, he felt no necessity for the presence of idols or temples. The only 
temple required by such a creed was the sacrificial floor, and the air itself sufficed for 
the sanctuary of the Devas or " bright beings" who ruled the several departments of 
the universe. An altar was erected under the open sky, wine was freely poured out, 
butter was thrown on the mounting flame ; and while the invocations were being 
recited and the flesh of the victim roasted, the unseen deities of the air were invited to 
come down and lick the blood, inhale the fumes of wine, and taste the savoury smoke 
that ascended from the altar. No priestly caste was needed for such a simple ceremony; 
and there is abundant evidence of sacrifices having been performed and invocations 
composed and uttered by the military chiefs, who led their people out to the field of 
battle. Gradually, however, when the warlike instinct had begun to yield to the liter- 
ary and religious, and the hymns were collected into liturgies, the art of sacrifice became 
much more highly developed than before. As time went on, it became more and more 
difficult, and at last impossible, for the king or his ministers to master the elaborate 
procedure which the Devas had now begun to demand ; and as the slightest error in the 
performance was believed to be fatal to its efficacy, a class of men came into existence 
who made a special study of the prescribed rituals and transmitted such knowledge to 
their posterity. As Brahma (masculine; meant sacrificer in general, without any dis- 
tinction of tribe or class, so (by the well-known rule of etymology wifh which every- 
Sanskrit scholar is familiar), the word Br&hman a meant the son or descendant of a Brah- 
ma—that is, one who had inherited a knowledge of the sacrificial art from bis fathers 
and forefathers. This was the origin of the word Brahman, and in this way the 
nucleus of the Brahmanical caste was formed. Thus the Brahman is no exception to but 
rather the strongest verification of, the theory maintained and exemplified throughout 
this paper, that function, and function only, was the foundation of Indian caste. 

107. It is not surprising that an exclusive priestly caste should have thus been 
gradually formed, when we reflect on the extraordinary effects ascribed by the Hindus 
of that age to the power of sacrifice which Brahmahs, and Brabmans only, could call 
into action. The sacrifice was the food of the gods, and as the gods were the powers 
of nature personified, it was the sacrifice which sustained the fabric of the universe. 
In the Atharva Veda even the sacrificial ladles were said to support heaven and earth: 
" the Juhu has established the sky, the Upabrit the atmosphere, and the Dhruva the 
stable earth" (Atharva Veda, XVIII, 4, 5); and similar powers are ascribed to 
the sacrificial grass (kusha) and to the sacrificial cow (Vasi). The, exclamation 
" Sv&ha,!" uttered by the priest as the homa offering was thrown on the fire was oon r 
verted into a goddesa and declared to be the wife of Agni, the consecrated flame of 
the altar. In a later production (ManuVCode; the sacvifite is said to " support the 
whole animal and vegetable wodd, since the oblation of clarified battel) duly cast into 
the flams ascends in smoke to tho son.; from the &ttn, it falls in rain; from rain comes 

Digitized by 


( 51 ) 

vegetable food ; and from such food animals derive their subsistence" (Chapter III, 
76). In a "still later production, the Bhagavat Git& or divine song, the power of sacri- 
fice in described in the following terms: " Prajapati of old created beings with their 
rites of sacrifice, and said — " Hereby shall you propagate yourselves; this shall be to you 
the cow of plenty. Sustain with this the god and the gods will sustain you.* support- 
ing each other in turn, you shall obtain the highest happiness. Fed with sacrifice, 
the gods shall give you the food that you desire. He that gives them nothing and 
eats the food whioh they give is a thief indeed. Good men, who eat the leavings of 
the sacrifice, are loosed from their guift ; but they that eoek for themselves alone, and 
not for the gods, eat sin." It is no wonder, then, that the Brahman, through whom 
alone the sacrifice could be made to yield its prondised fruits, was looked upon as a 
being possessed of superhuman powers and as one who held the keys of life and death 
in his hands — that every one demanded his blessing as he passed — that his curse 
was believed to penetrate to endless distances in time and space— that " no greater 
crime" (to use the language of Manu, Chapter VIII, 381) was known on earth u than 
the slaying of a Brahman" — that allother castes and classes were declared to have 
been made to protect or serve him— and that at the close of the long struggle for 
supremacy between the king and the priest, the letter was able at last to establish 
a theocracy more powerful and more lasting than any other that the world has yet 
witnessed. The final victory of the priest over the warrior was signalized in Indian 
legend by the slaughter of the whole Kshatriya race by Parasu R&ma, the armed 
champion of the B rah mans. 

108. The art of sacrifice received its highest development in the sacerdotal 
tracts or liturgies called Brahman&s, whioh were oompiled by and for Brahmens as 
guides to the performance of the great sacrificial ceremonies. The three orders of 
priests, answering to the three Vedas and their respective Brahmanas, were the Adh- 
varyu, who did the manual work connected with the sacrifice as prescribed in the 
Yajur-Veda; the Udgitri, who chaunted the sacrificial prayers collected in the Sima 
V$da; and the Hotri, or follower of the Rig Veda, who recited, not ohannted, the 
appointed prayers in strict accordance with the difficult rules of accentuation and pro- 
nunciation. The slightest error or omission in the performance required an expiatory 
rite — a principle observed, but to a less degree, in the monasteries of Europe at the pre- 
sent day, where a mistake in the singing of any part of the divine office entails upon the 
offender the ceremony of kneeling down before his superior in a conspicuous place as a 
form of penance or atonement. To prevent such mistakes and to superintend the other 
threeorders of priests, a fourth order was established with a fourth Veda (the Atharvan), 
and this functionary was called the Brahman proper. Thus the name Brahman came 
to be considered the highest title of priest ; and when the names of the other three 
orders died out with the disappearance of the Vedio sacrifices, it was the Brahman who 
gave his name to the entire fraternity. Thus in the code of Manu the only word used 
for a man of the saoerdotal caste is Brahman. 

109. There was one more development in the signification of the word Brahman 
which must be noticed in passing. It was laid down in Manu's Code that a Brahman's 
life should be parcelled out into four distinct periods or stages : firstly, that of a reli- 
gious student or Brahmach&ri ; secondly, that of a married householder or Grihastha; 
thirdly, that of a forest hermit or Banaprastha ; fourthly, that of an ascetic or Sannydsi. 
In the last two stages the recluse was directed to discontinue the sacrifices performed 
during the preceding period and devote all his intellectual energies to the study of 
Brahma, the Supreme Spirit. The books in which these speculations have been 
recorded are called (Tpanishads, and these treatises close what is known to Hindus as 
Sruti or the canon of Vedic revelation. Hence some native commentators have said 
that the Veda oonsists of two parts— one teaching Yagya, or the art of sacrifice, and 
the other Brahma, or the mystery erf the creative spirit. Now " Brahma" (neuter), 
as we showed in para* 106, was one of the names for a sacrificial prayer. It is derived 
from the root Mb, which signifies to grow, to expand. As a prayer offered in sacrifice 
was believed to transcend the boaods of space and support the whole animal and 

Digitized by 


( 52 ) 

vegetable world, so in the speculations about the origin of th* world the same word was 
used to signify the cosmic soul, the world-evolving spirit, the plastio power* that causes 
all the changes and processes of nature, being itself causeless and unchangeable. The 
contemplation and study of this Supreme Spirit, Brahma, gave a higher mean i rig to the 
word Brahman than that of a mere sacrifices " Whoever looks for Brahmahood else- 
where than in the Divine Spirit should be abandoned by the Brahmans" (quoted in Mass 
Mailer's Sanskrit Literature, page 23;. Elsewhere it is said that the visionary sage is 
alone the true Brahman : " Whatoyer kind of Brahman he may have before been, he 
becomes a veritable Brahman now" (OougVs Dpanishads, page 162;. The following 
sloka is taken from a commentary on Manu's Code ;— 

Janmand jayate Sddrah, sanshdrdd Dwijayo bhacet, 
Vedapdthir bhaved Bipro, Bralimajndndd Brdhmanah. 

The meaning of which is that by natural birth a man is merely a Sudra (that is, 
unregenerated and corrupt) ; by the ceremony of initiation he is regenerated and 
becomes a " twice-born ;" by reading the Vedas he becomes a partial Brahman ; 
but only by a knowledge of the Supreme Spirit, Brahma, does he become a full 



110. Having thus briefly shown the origin of the word Brahman and described 
the functions which first brought the caste into existenoe, we can now turn to the 
modern Brahman, with whom alone in this place we are immediately concerned. As 
the object of this paper is to describe the functions of the different oastes and to classify 
the oastes and sub-castes accordingly, I shall attempt in the following remarks 
to apply the same principle to the Brahman caste also — a task which is rendered the 
more difficult since, so far as I know, no systematic account has ever before been 
given or attempted of the functions practised by Brahmans at the present day. The 
genealogy and sub-divisions of the Brahmanical tribes so called have been described in 
great detail by the late Mr. Sherring in Hindu Tribes and Castes, Vol. I. He has 
shown how all Brahmans are supposed to have sprung from the seven great Rishis or 
sages of the Vedic age ; how 56 different gotras or orders are said to have sprung ou 
of theBe sages ; how the said orders are classified according to the Veda (Rig, S&ma, 
Yajur, or Atharvan), which they are supposed to follow ; how Brahmans are further 
sub-divided into kuls or clans; and lastly, how the clans are sub-divided into shdkhds or 
branches. His account, which covers 113 pages, is made up chiefly of lists of proper 
names. As this subject has been so fully treated already, I propose to leave it alto- 
gether alone and to confine myself to the object with which alone this paper is con- 
cerned—a description of the functions of the modern Brahman. I hold, too, that the 
elaborate genealogical scheme described by Mr. Sherring and accepted by Brah- 
mans themselves contains little, if any, historical truth, its very foundation — the seven 
sages — being mythical. Moreover, there are many sub-castes of Brahmans whose pedi- 
gree as Brahmans is said to be spurious, but whom it is necessary to inclnde as 
members of the great Brahman fraternity, because they are regarded as such by the 
general Hindu community. 

111. The following scheme shows how the Brahmans of the present day may be 
classified according to function. Some of these names represent separate castes or 
sub-castes— that is, distinct social* units — which never marry or give their daughters 
in marriage outside their own ranks ; others represent hereditary titles which must 
once have been names of separate castes, but are no longer so at the present time; 
others represent neither separate castes nor hereditary titles, but functions only : — 

f Hotri — Saorificer (Brahmanas and Kalpa Sutras) KJ ~ 
J BiduA— Consecrator of idols (Karcpn Kdnd and Sdma Ved): 
j • Aohfirya— Superintendent e* "ceremonies (Kalpa Sutras, DfofP Ska§ ~ 
tras f 8fc ) 
Dikshit -Initiator of t* e **ice-born (Karam Kdnd). \ 

,Pathak-Private tut<> r of <«tto (Siastras or sciences). \ 


\ ^\ 

< 53 ) 

fjyotishi— Astrologer (Jyotish, astronomy and mathematics). 
J Paurinik— Reciter of ancient histories (Purans and epics). 
II j Purohit— Family priest (Grihya Sutras aud Karam Kind). 
( PAnde— Village schoolmaster ( Vernacular directs and Ganti.) 
fOjba — Sorcerer-*- (Tantra and Kalika Purdna). 
I Panda— »Temple priest (Dash Karam). 
. . ^ Gangaputra— River priest (Dash Karam). 
j Joshi— Fortune-teller (Simudrika). 
|^Mah&- Brahman — Funeral priest {Garur Purina). 

In the previous groups ef castes I have mentioned the lowest first and the highest 
last. In the present group, however, I have reversed the order, because it was more con- 
venient to begin with the ancient or Vedic Brahman and make a gradual descent to 
those who are further removed from this standard, but represent most nearlv the 
religious life of the present day. Against each kind of Brahman I have mentioned 
the names of some of the books with which his function is specially connected. 
But it must not be supposed that all or even most of the Brahmans concerned are really 
acquainted with their respective books or with any portion of them. Except in a few 
places of exceptional note, like Ayodhya, Benares, Mathura, &c., Sanskrit is becoming 
more and more obsolete in Upper India, and most of the Brahmans named under 
heading III are now totally illiterate. 

112. The Hotri is the only class of Brahman still left whose title and function 
recall the animal sacrifices of the Yedic age, and even this function is rapidly be- 
coming obsolete in Upper India. The sacrifices of horses and cows, so famous in 
the Yedas and the great Hindu epics, are among the five things which have been declared 
unlawful in the KM yuga, or present age of the world. This is distinctly laid down 
in the following sloka taken from Par&sara's Code :— 

Aswamedham, gavdm lambhyam, sannyistham, pala-paitrikam, 
Devardt sutotpattim, kalau paneha bibarjayet* 

" There are five things which a man should avoid in the Kali-yuga — the horse sacri- 
fice, the cow sacrifice, the ascetic stage of life, the flesh offering to the eouls of the dead, 
and the raising of a son (to a deceased elder brother) from a younger brother." As the 
cow and the horse can no longer be used as victims, the modern Hotri is reduced to 
the less pretentious sacrifice of a goat ; and even when this is performed, it is seldom 
now done according to the Yedic rites and ceremonies. The last instance of a Vedic 
goat -sacrifice was performed in Benares about twenty-six years ago. The account 
which I have received of it k as follows : — 

A certain Brahman from Southern India, who had inherited the charge of an 
eternal fire from a remote and distinguished ancestry, had let the fire go out In 
order to relight it and make a suitable atonement for the crime he had committed, 
he determined to celebrate a great sacrifice, and collected from among all the Hindu . 
j&jas who could be induced to contribute a sum of about Rs. 30,000 for the purpose. 
The ceremony lasted for twenty-one days. All this time Brahmans were being fed in 
thousands, offerings of homa were being thrown upon the altar, and mantras or sacred 
words were repeated with each offering, The goat intended for the sacrifice which 
was to crown the work was stalled in an enclosure set apart as the sacrificial floor ; 
and the greatest attention, amounting almost to worship, was paid to it till the day 
of sacrifice came round. As the goat was being led up to the altar, its neck was gar- 
landed with flowers, and red powder was showered on its head. The most learned 
and distinguished Brahmans who could be found were summoned from hundreds of 
miles round to take part in this Yedic sacrifice. The spot on which the goat was at 
last killed for the immolation was screened off, so that no profane eyes might behold 
what the Brahmans were doing or witness the relighting of the extinguished fire 
from the flame of the sacrifice. On receiving his fire relighted, the man was taken to 
the Ganges to be bathed by the Ach&ryaor presiding priest ; and such was the sanctity 
ascribed to the water in which he bad been washed that almost the whole city of 


Digitized by V^OOQlC 

( W ) 

Benares turned out to get a drop or two thrown at them by the hands af the priest. 
It is said that no such sacrifice had been performed before within the memory of any 
man living, nor is it expected that such will ever be performed again. 

The substance of the homa offering which has given rise to the name of Hotri 
is made of ghf, rice, barley, til or oilseed, raisins, cocoanut, &c. r all mixed np to- 
gether. In large, sacrifices such as the above, or at times when some other religious- 
ceremony is performed with more than usual solemnity and cost, many maunds- 
of this sacrificial matter are collected, and priests succeed each other in turns throw- 
ing dribblets of it on the flame and reciting texts from the Vedae. At the time when? 
the homa is thrown on the fire twigs from certain trees or plants are thrown in witb 
it, — viz.. from the pipalor fig-tree, the palasa, the mango,, the catechu, the thorn-apple, 
the kasha gras?, and the banyan. These six are collectively called samidh. 

There is one more function left to the Botri which may be traced to the Vedic* 
age — the recitation of long passages from the Vedie hymns. This is done at time* 
when new temples are opened, or when large feasts are given to Brahmans,. and some- 
times in the private houses of rich men. A Hotri is sometimes employed by men,, 
who are wealthy enough to engage his services, to stand before an idol and recite 
extracts from the Vedas on their behalf. Such repetition is- believed to benefit the- 
soul of the man who pays for it* 

113. The Bidua is the class of priest specially employed to consecrate images r 
wells, tanks, and mango orchards. His name is derived from via\ ' 'to know," and» 
appears to be a corruption of the word vidyd, an ancient synonym for Veda. The 
point in common between the Bidud and the Hotri is that both have the reputation of 
being acquainted with portions of the ancient Vedas, and hence the one is sometimes 
employed for the same purposes as the other. The original functions, however, are dis- 
tinct. The Hotri performs or professes to perform what still remains of the old Vedic 
sacrifice, while the Bidui consecrates images and idols, to which the religion of the- 
Vedic age was an entire stranger. 

An image or any other symbol, such as a.lingam, intended to represent the pre- 
sence of a deity, is worth nothing more than the material it is made of, until it has 
undergone the process of consecration. The effect of this ceremony is to draw the 
deity down into the image or symbol, and to keep him there, so that henceforth the- 
image and the deity become virtually one; for "the Aryan brother" does not make* 
the distinction between the visible symbol and the unseen divinity, with which hi* 
admirers have sometimes credited him. The manner of the consecration is much the 
same for all Hindu deities alike, but the number of days spent on its performance 
depends upon the degree of dignity which the deity is supposed to possess. The idol 
©r other symbol intended for consecration undergoes various forms of ablution. First 
it is bathed in water taken from some sacred river, as the Ganges, the Narbada, the 
Sarju, &c, or any of these combined, if the water can be procured. Then it is bathed 
in the panchamrita, or " five drinks of immortality," w*., milk, cream, melted butter, 
honey, and sugar dissolved in holy water, first in each of these liquids separately, and* 
afterwards in the five combined. The image is then finally washed once more h* 
sacred water. The niche in the wall, or the block in the middle of the temple, on which 
the image or symbol is made to stand, undergoes the same ablutions as those applied 
to the idol. When the idol and its standing-place have been duly consecrated, the 
temple is ipso facto consecrated also* 

Wells, tanks, and orchards are consecrated in much the same way as idols. No- 
one is allowed to bathe in the tank, or drink water from the well, or eat of t the fruit of 
the orchard, till the above liquids have been thrown into them. It is chiefly in the- 
case of wells and tanks, which have been constructed from motives of public benefac- 
tion, that these rites of consecration are performed. 

An essential part of the ceremony in each case consists in feeding a horde of 
Brahmans and in making a homa offering to the gods similar to that already described 
in paragraph 112. At such times a vast number of deities or unseen guests are invited 

Digitized by 


( 55 ) 

to attend— (1) the nine planets (including the sun and moon) called Navagrah* ; (2} the 
twenty-seven lunar mansions called the Nakshatras ; (3) the constellation of the seven 
Rishis or sages called Sapta Rishi ; (4) the three hundred and thirty million deities who 
make up the vast pantheon of Hinduism ; (5) the Pitris or souls of departed ancestors, 
who make up another vast multitude almost equal to that of the gods ; (6) the ten 
Digpals who are said to preside over the ten points of the compass ; and (7; the sacred 
rivers of India and of the celestial firmament. For each of these seven groups a 
separate place, and a separate dish, very small in amount, are assigned, and they are 
invited to come down and taste, or at least smell, the offering made to them. The 
smallness of the feast thus prepared for the gods presents a strange contrast to the 
huge viands placed hefore the hungry Brahmans, who are regaled with a fresh meal 
every day as long as the ceremony of consecration lasts. 

114. The Achdrya (or as the name is now commonly spelt, Acharja) is the 
highest kind of priest in modern India. His place in the religious ceremonies of the 
present day is similar to that once held by the Brahman proper in the celebration of 
the Vedic sacrifices (see para. 108). The special function of the Acharja is to 
guide and superintend the Hotri and the BiduA, and his name is derived from 
dchdra, which means " rule or direction," If any great reHgious ceremony is to be 
held in which Hotris, Bidu&s, and others are required, there must be one man to preside 
and give the necessary directions, and this man is called Acharya. His supervision rs 
specially needed during the constant repetition of the koma offerings and the recita- 
tion of the appropriate Vedio teats. It is he, and he alone, who knows how to sum- 
mon the hosts of divinities who are invited to partake of the offering, and how to send 
them back into the sky contented and propitiated* 

There are very fewBrabmans at the present day whose attainments in Sanskrit are 
sufficiently varied and accurate to enable them to discharge the office of Ach&rys. The 
highest title which can be given to a Brahman is to call him an Ach&rya, and this is the 
title which has been selected by the Educational Department in these provinces to be given 
to those students who pass the most difficult examination in the Benares Sanskrit College. 

115. The priest specially employed to initiate a Hindu boy into the performance 
of his religious duties and to give him " the second birth" is called a Dikshit. The 
word is] simply a contraction of Dikshitri or Dikshitd, which signifies " one who 
initiates"; and hence it does not mean " initiated/' as Mr. Sherring and others have 
supposed (Hindu Tribes and Castes, Vol. I, page 9). It is only boys of the upper 
castes (that is, those who are called the twice-born) who are entitled to the privilege 
of diksha. But Brahmanism has for the last two hundred years and more been steadily 

decending into lower and lower strata of the population, absorbing one indigenous tribe 
after another ; and hence the possession of this privilege cannot now be considered a 
mark of " twice-born" ancestry. 

The orthodox age for undergoing the rite of diksha or initiation is on the comple- 
tion of the seventh year. The Hindu book of ceremonies known as karam kdnd calls it 
the eighth, but the figure is raised to eight by counting the nine months preceding 
birth as an additional year. At the present day the orthodox age is not always observed, 
and a boy can be initiated a year or two later, if it suits the convenience of the parents 
to postpone incurring the expenditure which these rites^ entail. A boy, whatever his 
parentage may be, is not a full Hindu until the diksha has been performed. Up till 
then he is little better than a Sndra or unregenerated person. But on and after that 
day he incurs the religious responsibilities to which his parents have all along intend- 
ed to dedicate him, as a Christian boy does by the double rite of baptism and confirma- 
tion. Girls are never " initiated" as boys are ; and thus a high-caste woman who 
marries a man of the Sudra rank cannotbut become a Sudra herself. This, I suspect, 
is the real explanation of the abhorrence felt by Hindus to a woman being married into 
a caste lower than her own. The same abhorrence has never been felt to a twice-born 
man marrying or cohabiting with a Sudra woman, for the woman can rise to the rank of 
her husband ; but as she has never been initiated, she cannot raise the husband to her 

Digitized by 


< 56 > 

own. Thus in Mann's Code a Brahman was allowed to take a Sudra woman into his 
house ; but if a Sudra man married a Brahman woman, the son became a Chandal— 
" a sinful and abominable wretch." 

The entire ceremony of diksha lasts some eight or nine days. Throughout these 
days the boy is put upon a very strict diet and undergoes a rigorous course of ablutions, 
fie is bathed regularly at certain hours; after the bath mustard powder and oil are 
rubbed on his body ; and he then undergoes a second bath to wash them off again. 
All this time he should wear nothing by day or night but a string made of the sacred 
grass called kusha, which is tied round his waist, and to which a narrow cloth called 
langoti is attached, fastened between his legs before and behind. Meanwhile the usual 
homa offerings are thrown on eon sec rated fire by priests of the Hotri class, who have 
been summoned for this purpose. When the last and greatest of the homa offerings 
has been made, the sacred thread (upavit or janeo) is thrown over the left shoulder of 
the boy by the Dikshit, and the first act of the ceremony of initiation is com* 
pleted. The Dikshit then throws a cloth over his own and the boy's head, and under 
cover of this cloth be instils into his ears (in an under tone, so that no profane ears 
may catch what he says) the Gayatri and all the other sacred verses which a Hindu 
should utter on stated occasions every day of his life. The repetition of all these 
verses, and especially of the Gayatri, which is repeated first, constitutes the closing cere- 
mony by which the boy is formally initiated into the rites of Hinduism. The boy must 
have heard and seen something of these rites beforehand through living with his 
parents; but until he has been formally initiated, and this by a Brahman competent 
to discharge the office, he is a mere heathen. For some weeks after the conclusion 
of the ceremony the Dikshit remains with the novitiate, so as to help him to perform 
the several daily rites and make him sufficiently perfect to be left to himself; and after 
leaving him, he continues to be his spiritual adviser for the rest of his life, whenever 
such advice may be required. 

There is no space to enter minutely into the rites which make up the daily 
religious life of a Hindu, and for the teaching of which the Dikshit priest is specially 
employed. Certain verses have to be repeated and certain ceremonies performed every 
day before eating, drinking, and sleeping. In fact there is scarcely anything that a 
Hindu can do but he finds the cloud or sunshine of religion darkening or illuminating 
his path. The most important, perhaps, of the daily rites are the morning, evening, and 
midday oblations, which are collectively called the Trikal SandhyA. Every morning 
in his life a man must rise before sunrise and bathe. While the sun is still half visible 
above the horizon, he must re-enter the river or tank and first throw up an offering 
of water with the palms of his hands (both palms being joined together) to the gods, 
the sages, and the souls of ancestors, and then make a final offering of water to the 
sun (arka). After making the offering to the sun, he must repeat the Gayatri one 
hundred and eight times. The same ceremony is performed with very slight changes 
at midday and again in the evening. Each ceremony is called Sandhya, or the 
junction ; sunrise and sunset being the hours when day and night meet, and midday 
the hour when the easterly sun meets the western. 

116. Priests who devote themselves to the teaching of the young have beetf 
called P&thak or Upfcdhay, the former name being best known in Upper India and the 
latter in the Lower Provinces. Both words mean literally "teacher." 

The function of the P&thak begin* where that of the Dikshit or initiator ends. 
In the old Hindu discipline the life of a twice-born or initiated person was parcelled 
out, as we showed in a former paragraph, into four successive stages or avasthas 
The first of these was that^of student or Brahmach&rj, and this was spent under the 
care of the P&thak or Up&dhay. The Pathak, then, was a very important functionary 
in the days when Hinduism was strong and when the people were strictly ortho- 
dox. It devolved upon him to train the character of the young after the Brahmanical 
model and lay the foundations of a religious life. He instructed his pupils partly in 
the Vedas or religious poetry, but chiefly in the Shastras or so-called books of science, 
The office of P&thak, like that of almost every functionary in India, from the sweeper 

Digitized by 


( 57 ) 

to the priest, is often hereditary among a certain circle of families. The sons of his 
constituents go to his house to receive daily instruction, and the sons of poorer Brah- 
mans, who have no P&thak of their own, are encouraged to attend with them. All 
tuition is nominally gratis. In fact the Pfithak is not only debarred from receiving 
monthly tuitional fees, but he is even expected to feed and maintain his pupils for 
nothing. They in return perform many kinds ,of menial offices for him, such as 
washing his feet, spreading his bed, drawing water from the well, driving his cows 
out to pasture, milking them when they come in, &c, nor do they disdain to eat the 
leavings which come from his table. There are two different ways in which a P&thak 
is maintained. Sometimes a r&ja or other rich man provides him with land, the rent 
of which keeps him above want and enables him to feed his pupils. Such endowments 
are not uncommon at the present time in Upper India, and in some cases such lands 
are exempted from paying rent or tax to the British Government. Another way in 
which the Pathak is maintained is by donations, called guru dakshana, which are given 
him by his wealthier pupils after they have completed their education and come into 
their patrimonial estates. The custom is for the ex-pupil to go to his former teacher 
and ask him what he would like to have, and whatever the teacher may say the 
ex-pupil is bound in honor to give. 

117. The five kinds of functions described thus far, viz., those of the Hotri, 
Bidu&, Acharja, Dikshit, and Pathak, are the highest and most respected in which 
a Brahman can engage. It is only the " twice-born" classes or castes who have 
occasion to employ Brahmans for such purposes, and this is the main reason why the 
functions corresponding are held in such high repute. The case is different, however, 
with the functions which we have now to describe. These stand on a lower level than 
the preceding, because they are not less frequently exercised on behalf of castes or 
classes of the population, who are 'below the rank of " twice-born." The principle of 
" condescending to men of low estate," which constitutes the ideal of the Christian 
and Buddhist creeds, is entirely alien to the spirit of Brahmanism and was strictly 
prohibited by Brahman law-givers themselves (see Manu's Code, Chapter III, 65, and 
IV, 61, 61, 99). A Brahman who does anything to help or enlighten a low-caste man 
lowers himself by so doing, and the only motive that can lead him to commit such an 
impropriety is the fee which he exacts in return. Nevertheless, the practice of offici- 
ating for low-caste men is constantly on the increase, and such Brahmans now make 
up the bulk of the Indian priesthood. 

118. The Jyotishi is a Brahman specially versed in astrology. This function 
consists in interpreting the will of the stars to his clients. His name is derived from 

jyotith, the Vedic name for astronomy, a science which formed one of the six Vedangas, 
or branches of Vedic exegesis. 

The object for which astronomy was studied in ancient India, and which gave 
the first impulse to such studies, differed essentially from the uses to which such 
knowledge is now applied. Its sole purpose in the Vedic age was to fix the days and 
hours of the great periodic sacrifices. In India, as elsewhere, the earliest mode of 
measuring time was by the changes of the moon ; and the asterisms lying in the 
moon's path were called Nakshatras or lunar mansions. Even in the earliest hymns 
these asterisms have not only received proper names but have been personified, and 
are invoked as deities to grant progeny and other blessings to their worshippers. 
In the first book of the Rig Veda mention is made of the thirteenth or intercalary 
month, and in certain hymns of the Sajur Veda this month is invoked as a deity 
under the name of Anhasaspati. In the later epic poems the Nakshatras are 
declared to be the wives of the moon-god, and the daughters of Daksha, one of the 
old names for the Creator. The Vedic name for astronomer was Nakshatra-darsa, 
" one who studied and observed the lunar mansions," and fixed thereby the date* 
of the periodic sacrifices. 

The modern name for astronomer is jyotishi. After the old system of Vedic sacri- 
fices had died out, the functions of the astronomer took an entirely different turn. 


Digitized by 


( 58 ) 

He became dn astrologer rather than an astronomer, and this is the only, light in 
which he can be now regarded. The Nakshatras are still, as before, looked upon as 
divine persons. Bat a mach greater degree of prominence has been given to the 
influence which they are supposed to exercise on the events of human life. Every 
Hindu who is not an infidel (and total loss of faith is rare in Upper India, even 
among men educated in European sciences) believes that his fate is written in the 
stars ; and as the moon and stars must inevitably run their courses, it is no wonder 
that Hindus are firm believers in fatalism. But to every evil influence which a star 
is expected to inflict there is a shdnti or propitiatory ceremony attached, the object of 
which is to diminish, if not to avert, the calamity; and it rests with the astrologer to- 
find out what evil the stars are about to inflict and to apply the remedy. The con- 
sulting of the Nakshatras plays as important a part in the life of a Hindu as even the 
institution of caste, and nothing serious can be done or undertaken till the stars 
have been declared favourable. 

After the birth of a child, the first thing that the father does is to go to the jyoti- 
shi and tell him as exactly he can the hour in which the child was born. The jyotishi 
then consults the stars and casts the horoscope, from which the fate in store for the child 
is determined. If the astrologer's forecast turns out to be wrong, and he is twitted by 
the parent for having proved a false prophet, the astrologer taxes him in return with 
not having told him the hour of birth correctly, a pretext which can be easily set up 
in a country where clocks and watches are not in common use. At times of sickness 
or any other calamity the astrologer is consulted as to whether there is any evil star 
in the ascendant by which the calamity is caused. If the answer is in the affirmative 
(as. of course it always is), the man seeking advice is told to make some offering in cash 
or kind in order to appease the hostile star; and as the astrologer is the recognised 
exponent of the starts feelings and wishes, he appropriates, and is expected to appro- 
priate, the propitiatory offering. Thus the offering made to the star becomes part of 
the astrologer's fee. Sometimes, if a man is applying for a situation or for a higher 
salary, or seeks to secure success in some particular undertaking, a bribe is offered to 
the star (through the astrologer with whom the star is in league) to help him to gain 
his end. Not being able to whip the stars, as some nations have whipped their frogs 
in order to hasten the fall of the tropical rains, they use the astrologer as the medium 
for bribing tbem. * 

For settling betrothals and for the performance of marriage ceremonies the 
services of the astrologer are indispensable. When the family barber or Napit has 
selected a boy whom he considers a suitable match for some girl of the same caste 
(see above para. 97), no contract can be made between the parents till the astro- 
loger has been consulted as to whether the stars of the boy are not inimical to those 
of the girl. Nor is this even enough. He must also find out what the castes of the 
boy and the girl were in the former state of existence. If both were of the same caste 
in this pre-existent state, the betrothal contract can be made, so long as the stars are 
not otherwise hostile. But if it turns out that the caste of the boy in a former state 
was below that of the girl, the betrothal is disallowed. Brahmans have framed all 
these rules entirely to their own advantage. A man who was a Brahman in a former 
state may marry any girl who in a former state belonged to some caste below his 
own, but the rule is never reversed. Some idea may be formed of the obstructions 
which can be thrown in the way of marriage amongst Hindus, when it is known that 
compatibility of caste in a former state of existence is only one out of the thirty-six 
conditions that must be complied with before a betrothal can be declared valid. The 
most enlightened Hindus of the present day are generally compelled to comply with 
these conditions, however much they may desire to discard them. 

An auspicious day must be selected by the astrologer for almost e very thin <r that 
can happen in a man's or woman's life— for the day of marriage, for every part of the 
marriage ceremony, for starting on a journey, for putting the first plough to the 
soil, Ao. A woman cannot put on a new set of bangles until she has learnt that the 
stars are favourable, and an orthodox man will not put on a new garment unless he k 

Digitized by 


( 59 ) 

assured that the day is lucky for first wearing it. It is matter of history that when 
Ala-ud-din Kbilji was marchiug upon Nadiya, the great centre of Brahmanism in 
Bengal, the Hindu raja was ordered to wait for nine hours before marching out to 
meet him, because until then the stars were declared to be unfavourable. Meanwhile 
the Musalman forces entered the city, and the raja had to flee for his life. On the 
other hand, there is a Hindu proverb to the effect that necessity has no laws, even 
though the stars may be against it: — Ghari men ghnrjare, Arhdi ghari bhadrcu That 
is, " if the house will catch fire in twenty-four minutes, an hour (hence) is an 
unlucky moment for action." 

In the constitution of the Hindu township the astrologer is a most important 
functionary, and men of all castes, whatever their rank may be in the social scale, look 
to him for the interpretation of the stars in every domestic event or industrial under- 
taking. As the jyotishi does not confine himself to the " twice-born" castes, but 
officiates for men of low estate, provided they can pay him, he has been and is one of 
the chief means by which the casteless tribes have been brought within the pale of 
Hinduism; and this is a process which is continually going on before our eyes. The 
Hotriand Bidua* cannot recite Vedas, or perform the homa sacrifice, or consecrate idols, 
for any but the " twice-born" castes ; and it is only in those families which have 
established a similar title to the rank of twice-born, that the Dikshit (initiator) or the 
Pathak (private tutor) will consent to serve. But the astrologer does not refuse his 
services to any man who offers him a fee proportioned to his means ; and among the 
various functions that coma within the sphere of a modern Hindu priest there is none 
that appeals more directly to the minds of ignorant and superstitious men. It seems, 
indeed, that belief in the influence of the stars upon human destiny has been one of 
the universal instincts of mankind ; and the jyotishi comes armed with a reputation 
for mystic knowledge which to the simple mind of the savage or low-oaste man is at 
once convincing. One of the first symptoms of a savage tribe becoming Brahmanized 
is that they have begun to oonsult the astrologer. The asirologer, then, has played 
and is playing a very active part in drawing the indigenous or non- Brahmanized 
tribes within the net of Hinduism, and the intercourse which he is thus tempted to 
hold with the unregenerated masses has, as we have just shown, made his office appear 
less respectable in the eyes of other Brahmans. All castes, however, hold him in awe, 
and it is impossible to foresee the day which will witness their escape from his thral- 
dom. Muhammadans have almost as much faith in his pretensions as Hindus. 

119. The Pauranik, as his name implies, is one who makes it his calling to read 
out the Puranas, or ancient histories so called, in the presence of mixed audiences. 
Women as well as men attend such readings ; but if they belong to the upper castes 
they are screened off by a curtain. The manuscript or printed paper from which the 
reading is made is called a poti (a corruption from the Sanskrit pusiak) ; for nothing 
so profane as a book bound in leather is used on such occasions. The poti is worship- 
ped as a fetish. Before the reading is commenced, the man in whose house the enter- 
tainment is held bows before the poti, makes it an offering of rice, sandal- wood powder, 
flowers, &c., just as he would make before an idol ; and if be is a man of approved piety, 
be repeats this offering every morning, so long as the reading lasts. Even the priest 
who reads receives something like divine homage ; for his forehead is painted with 
sandal-wood powder and he is crowned, like an idol, with a chaplet of flowers. One 
or two hour3 are set apart every evening for the reading, and sometimes three months 
are spent before the poti is finished. At the close of the performance every member 
of the audience presents an offering to the poti, as to a divinity or idol — an offering 
of cash, or grain, or a piece of cloth, each giving according to his means. After 
these offerings have remained a little time in front of the poti, the Pauranik takes 
them up and appropriates them as his own fee— an arrangement which is recognised 
as perfectly correct and legitimate by the donors present. The reader is believed to 
impersonate, for the time being at least, not only the book, but the gods and demi-gods 
whose actions it records ; and as neither the book nor the gods appropriate the offer- 
ing, it is rightly made over to the priest who represents them. 

Digitized by 


( 60 ) 

It is chiefly in the rainy season, when there is less scope for occupation in the 
open air, that these readings are held. They take the same place in Hindu social life 
as that of the Sunday sermon or week-day lecture in Europe. No one but a Brahman 
can exercise this function. He may be a worse reader than a Kayasth, or a Chattri or 
even a common Kahar, but this matters nothing. Spoken by a layman the words lose 
their imputed sanctity ; spoken by a Brahman, they illuminate the soul of the listener, 
even if he understands little or 'nothing of their meaning. Pious men, who have the 
means, sometimes keep a PaurAnik to read to them every day of the year. 

There is one more fact deserving of notice in regard to the Paur&nik before we 
part with him. Like the jyotishi, he has been and is largely instrumental in convert- 
ing the indigenous tribes and ignorant classes of the population to Hinduism. To 
read anything sacred within the hearing of a Sudra, much more within that of an 
outcaste, is a practice against which Brahmans are cautioned in Manu's Code, on 
pain of incurring the most terrific consequences in the future life. But when the 
Brahman begins to weigh the wants of the present life against the threats or 
promises of an uncertain future, he not unfrequently decides in favor of the former, 
and the terrors of hell have proved far less effective than the attractions of pice. It 
is seldom that a camp of Kanjars or Nats, or other casteless and wandering tribes, 
can remain for several months together on the same spot, but some Brahman finds them 
out and opens his Purina and commences to read aloud before an ignorant and gaping 
audience. This is often the first step taken by an Indian savage towards entering 
within the fold of Hinduism. It is easy to conceive how in ancient times small roving 
bands of cattle-grazers, such as Ahirs, Gujars or Gaddis, and afterwards hunting 
bands of Pdsis, Bhars, Arakhs, and many more, were thus caught as it were by the 
hungry Brahman, and detained in the outskirts of the village, till they became part of 
the regular inhabitants and, having abandoned their own tribes, formed the nucleus of 
what are now known as the lower castes. It is thus that Kanjars, Doms, Tharus, &c, 
are now being drawn within the net, from which to those once caught there is ne escape. 

120. The office of Purohit or family priest was one of much distinction in ancient 
times ; but there is now very little dignity attaching to it. The name simply means 
" superintendent," " master of the ceremonies." It was one of the oldest names, proba- 
bly older even than "Brahman," for a professional priest. For it was the custom of 
kings in the Vedic age to employ some man, versed in the sacrificial art, to perform 
the sacrifices for them, and this functionary was called Purohit. The violent contests 
between Vasishta and Yisvamitra, two of the most distinguished prelates of the Vedic 
age, for the post of professional priest in the court of king Sudas, show how much 
importance was attached to the office in those days. 

The office of Purohit seems to have been hereditary from the first, and this is one 
of the chief reasons why it has become so insignificant in modern times, A man 
who is certain of his appointment and of being able to bequeath it to his son will not 
take the trouble to go through the severe course of training to which Brahmans 
desirous of rising in their profession are ordinarily subjected, and hence the family 
priest became as lazy and illiterate as we now find him. In these days a purohit 
can seldom discharge any but the most petty offices for his master, such as presenting 
the daily offering to the family gods, or performing the usual rite when the first plough 
is put into the soil or when the harvest is being brought in. On great occasions, such 
as the performance of a marriage ceremony, or the dedication of a temple, or the con- 
sultation of the stars, or the initiation of a son, the Acharya, or Bidui, or Jyotishi, or 
Dikshit are called in, although such offices would come well within the duties of a 
family priest, if he were competent to perform them. On these occasions, he renders 
what help he can under the guidance of the Achfirya or other invited priest, particu- 
larly in doing the preparatory or (as we should call it) the "dirty" work. At the 
appointed season he assists his master in paying the annual offering (sraddha) to the 
souls of the dead. If his master has been keeping a fast, the purohit is ready enough 
to act the part of the Brahman, who, according to Hindu rules, must be fed first before 

Digitized by 




( 61 ) 

the fast can be broken. If bis master has a private shrine in his house, he acts as the 
temple-priest, tending the lamps and sweeping the floor. If his master is seeking to 
get his daughter betrothed, he helps the barber to find oat some suitable boy. If his 
•master is taking a journey, the purohit sometimes goes with him to act as cook ; for 
no caste in India will refuse to take food cooked by a Brahman. If his master wishes 
to give a feast to JBrahmans on some periodical or other festival, the purohit invari- 
ably offers his services both as caterer and as consumer. At all religious ceremonies, 
should the master be unable to attend, the purohit can act as his proxy. He can go 
on a pilgrimage for his master to some distant shrine, fast for him at home, and even 
bathe for him in some sacred stream or tank. 

Every orthodox Hindu is glad to keep a purohit if he can afford the cost, just as 
the Israelites of old esteemed themselves fortunate if they had a Levite on their establish- 
ment. Every reader will remember the story of Micah, who, having met with a 
stray Levite, said unto him—" Dwell with me and be unto me a father and a priest," and 
duly installed him in the office. Then said Micah — "Now know I that the Lord will 
do me good, seeing that I have a Levite to be my priest " (Judges, xxvii,13). 

But the functions of purohit are not now limited, as they once were, to the twice- 
born castes. There are certain classes of Brahmans, calling themselves purohits, who 
have established priestly relations with the inferior tribes and castes, and visit them 
when the occasion arises, receiving gifts in return. There are, for example, Oham&rwa 
Brahmans, and even Dom Brahmans, who give certain help to the tribes corresponding 
at times of marriage, Ac. A savage, who has gone so far as to oonsult an astrologer or 
hear a Paur&nik read to him, will generally go a step further, if he remains long 
enough in the same place, and attach himself to some Brahman 4 who will act as his 
purohit or family priest whenever his services are required. When this stage has 
been reached, the captivity of the man is no longer a matter of doubt. Henoeforth he 
becomes a Hindu, attends the great public festivals, bathes in the holy waters, visits 
the sacred shrines, and, though he may not be allowed to enter a temple, he can employ 
the temple-priest to place his offering on the idol. It is thus that the indigenous tribes 
of India have one after another been drawn into the net of Hinduism, until, in Upper 
India at least, there are scarcely any more tribes to be caught. 

121. The P&nde (like the Pathak or Up&dhay desoribed in para. 1 16) is a teacher 
of the young, but of a much lower stamp. The name is derived from pandd, science 
or knowledge ; and hence Pandit has become the title of a Brahman deeply versed in 
Sanskrit. Perhaps, then, originally the functions of the Pinde were scarcely, if at all, 
inferior to those of the P&thak : but for some centuries past this has not been the case. 
The P&nde does not now teach Sanskrit, but only Hindi. His tuitional course is 
not merely religious, but secular also ; for it includes the elementary arithmetic (ganit) 
used in the native markets, and the quick running hand known as Kaithi. He is in 
fact " the village schoolmaster," and as such held an important place in the ranks 
of the Hindu township. His school was less exclusive than that of the P&thak ; for 
the latter received only twice-born pupils, most of whom were Brahmans ; while the 
former did not refuse admission to boys below the rank of twice-born, provided these 
were not of castes whom it was a pollution to touch. The Sanskrit patshalas, which 
we now see around us, and which are becoming fewer every year, are the legacy 
of P&thaks, while the Hindi patshalas which have survived in larger numbers 
are the legacy of Pdndes. The existence of these two classes of schools is the 
sole foundation for the assertion which has been made of late, that India possessed 
in former times a complete system of indigenous schools open to all classes of the 

Men of the Kayasth caste have now acquired a large share of the function onoe 
monopolised by P&ndes. Yet all such Eayasths are called gurus or spiritual guides, 
a title which, properly speaking, could be applied only to a Brahman. 


Digitized by 


( *» ) 

112. The Ojha Brahman is one who is specially versed in (he practice of spells 
tod charms. Most of these rites are described in the books called tantras ; and hence 
» Brahman who deals in them is sometimes called by the name of Tantrik, that is, 
wizard or sorcerer. It is difficult for outsiders to investigate the mysteries of Indian 
magic ; for the interpretation of the tantras is more esoteric than that of any other 
branch of Sanskrit literature ; and it is obviously to the interest of the sorcerer 
caste to keep their knowledge as much as possible to themselves. To this, more 
than to any other class of Brahman, the well-known couplet given below is 
applicable :— 

Devadhinam jagat sarvam, mantradhinaseha devatah. 
Te mantrah Brahtnanddhinah, tasma Brahmano dewata. 

" Th»*lti}e world is in the power of the gods, and the gods are in the power 
Of magic; magic Ifife- the power of the Brahman, and therefore the Brahman is him- 
self the god" t— a true desoHptfon of a priest who professes to control the higher powers 
by magical words and deeds andloTJ&mpel them to interfere in the affairs of men 
according to his own wishes. \^^ 

Hie word tantra is of Sanskrit origin and miter "" «t rin g ° r system" of cere- 
monial rites. But the name Ojha (by which the Brahwjiwizard » most commonly 
known) is derived from the word ojh, which signifies "enm&' > *nd w of a purely 
indigenous or non-Sanskrit source— indicating, what is the fact, tot the art of sorcery, 
though in a ruder and less cultivated form, is well known to those safeg* *nd casteless 
tribes of India which are Btill outside the pale of Brahmanism, and i£at Brahmans 
themselves acquired the art from the aboriginal races. The aboriginttk *oroerer is 
balled Ojha because he examines the entrails of the victim immediately mto* it has 
been slain, while the Brahman sorcerer does not In spite of this, hdWwer, the 
name Ojha has fastened itself upon the Brahman also. 

The office of sorcerer, within the Brahman caste at least, is striotly hef Gditay* 
and thus every man belonging to this class of Brahman is a potential 
a severe and prolonged probation must be undergone before a man is comi 
commence practice ; for he cannot get the deity into his power by the more 
hereditary claims. He must go through a course of path and jap 9 the former < 
consists in reading aloud, and the latter in silently repeating, the name of 
to be conquered a million times, more or less, according to the nature of the sp 
acquired. Special hours are assigned for this performance, special postures^ 
body, special diet, and a specially appointed space, beyond which the proba 
not allowed to go until the process of initiation is finished. 8pecial texts, 
associated with special spells, and spells have been classified under foe 
headings, which are as follows: — 

(1) Hdran, or putting a man to death. 

(2) Uchhdtan, or getting him out of the way without killing him. 

(3) Vashikaran, or getting him into one's power so as to use him as a tWJ>°k 

(4) Akarshan, or drawing him towards you from a distance. 

Under each of these main heads there is a variety of sub-headings why ' 1 n66( * 
not be here enumerated. Different deities are attached to different spells, lland it is 
though the medium of the deity attached to each spell that the sorcerer is |k believed 
to work. All of these deities, however, are at bottom really one ; for they aflfr° mere 
variations of Kali, the wife of Mahadev and the dreaded goddess of death. V * n & e 
Hindu pantheon there are ten main forms of this goddess besides several minefcr ones, 
and these ten forms are known collectively by the name of Dashmak£vidy6. p Each 
separate form, however, is treated as a distinct personality. The average sorcerer 
contents himself with captivating three or four only of these as his accomplices.^ Bmt 
men of exceptionally high teaming or ambition aim at acquiring a mastery everN4i* 

Digitized by 


( «3 ) 

entire group. Sferaswati, the goddess of learning and wife of BrahmA the Creator, 
and Lakshmi, the goddess of wealth and wife of Vishnu the Preserver, are some- 
times invoked by the sorcerer in conjunction with K&li, the wife of Mah&dev the 
Destroyer; but this is only done in oases where wisdom or wealth happen to be the 
objects specially sought for, and even then these benign goddesses lose something 
of their usual gentleness through being associated with such a hideous divinity as 
K61i. Those Hindus who worship the female powers (Sakti', to the neglect of the 
male triad known as Brahmd, Vishnu, and Mah&dev, are called S&ktyas, and to 
such men the tantras are the favourite scriptures, and the Ojha or Tantrik Brahman 
is the favourite priest. 

Brahmans of the Ojha caste are also called Panchmak&ri, because the conditions 
under which they perform their rites are expressed in five words, each of which 
begins with m : mdnsa, or eating flesh; madra, or drinking wine ; mantra, or repeating 
magical words ; mudrA, or putting the limbs and fingers in certain postures ; and 
maithun, or the association of the wife with the husband. As the sole object of wor- 
ship is a goddess, and as none but che female powers of creation are recognised by 
Brahmans of this class, the ceremonies are considered null and void, unless the wife 
of the priest takes part in them and repeats in act and word everything that he him- 
self does and says. If the priest happens to be far away from his house, or if for 
any reason his wife cannot accompany him r or if he happens to have no wife at the 
time when his services are wanted, he engages a prostitute and lives with her as her 
husband for such time as the ceremonies last. This arrangement answers the pur- 
pose equally well ; for the efficacy of the ceremonies is not impaired so long as some 
woman living in conjugal or ^uosi-oonjugal union with the man takes part in it. 

In some parts of India the female principle is worshipped, not merely by name 
and with the help of symbols, but in the person of the woman herself, and at some 
of the temples of Kffli, especially those in Assam and Bengal, promiscuous intercourse 
is said to form part of the orgies. 

There is scarcely any reason for doubt that Brahmans of the Tantrik or Ojha 
caste are to a large extent descended from aboriginal priests, specimens of whom 
abound even at the present day among the indigenous and un-Brahmanized tribes 
of Upper India, such as Doms, Thftrus, Eanjars, Nats, &c. Even among these tribes 
the function of sorcerer has a tendency to become hereditary ; but when such tribes 
become converted to Hinduism, the hereditary principle which Hinduism has so 
consistently enforced in every other case would be openly avowed and confirmed, 
and thus the aboriginal priest would naturally rise to the status of Brahman. The 
Ojha Brahman is so utterly unlike the Brahman of Mann's Code in manners and 
character, and so very like the Ojha of the aboriginal tribes from whom he has 
borrowed his name, that it is impossible to avoid the conclusion that the one is 
descended from the other* Both lean upon the same goddess, K&li ; both undergo a 
severe course of physical and mental discipline before she will consent to use them 
as the instruments of her power ; both indulge largely in flesh diet, in intoxicating 
liquor, and in free intercourse with women ; both shed the blood of animals before 
the idol which they worship ; both expel devils from the sick or drive them into 
those who are whole ; both use magic, spells, and charms. The difference between 
them lies in the fact that the Brahman sorcerer has reduced his art to a written code 
or system (tantra), while the aboriginal sorcerer has remained a coarse and illiterate 
savage. But even among those practitioners who from time immemorial have been 
credited with the name and rank of Brahman, and who are employed as such by 
the respectable castes of Hindus, there are some who are quite illiterate. Between 
these and the aboriginal priest no substantial difference exists. 

Almost all Brahmans of the Maithil tribe (as distinct from those of the Saraswat, 
Kanavjie, Jijhotiya, and Gaur tribes) practise the function of Ojha ; and \t would 

i • 

( 64 ) 

seem that Bengal had no Brahmans of any other stamp until the time of Adhiswar, 
king of Gaur, who lived in A.D. 900, and invited (as the legend says) five distin- 
guished priests from Kany& Kubja (Kanauj) to enlighten his people and himself in 
the milder creed of Brahman and Vishnn. These five priests became the founders of 
the great Kanaujia families in Bengal, and introduced a new class of rites and tenets 
distinct from those of the Maithil Brahmans, who trace their origin to Mithila or 

The late Mr* Sherring, after giving a brief account of the Ojha Brahmans of 
Benares (which, however, is more true of the aboriginal than of the Brahmauical 
sorcerer for whom it is intended) concludes with saying : " Formerly the Ojba was 
always a Brahman, bat bis profession has become so profitable that sharp, clever, 
shrewd men in all the Hindu castes have taken to it." This is quite incorrect. It 
implies that Brahmans were the first inventors of sorcery, and that the art has since 
been filtrating downwards from them to the castes immediately below, and from these 
again to castes still lower. The truth lies in the opposite direction, as we have al- 
ready explained. Moreover, it is wrong to say that Ojhas can be found " in all the 
Hindu castes." They are only to be found in the very lowest— that is, io those back- 
ward and despised communities whom I have placed in this paper in the class of 
hunting and fishing castes and in the ranks of the lower artisan and serving castes. 
As these tribes were the last to be brought within the pale of Hindu castes, their Hin- 
duism is still of the faintest possible type, and hence they have not yet altogether dis- 
carded men of their own tribes as priests or sorcerers. But as soon as a caste 
begins to rise in the social scale, that is, to make a nearer approach towards Hinduism 
proper — the aboriginal priest disappears, and Brahmans of the various orders and de- 
grees take the entire superintendence of divine matters into their own hands. No 
such thing as atn Ojha could be found or has ever been heard of among Chattris, 
Ehattris, Eayasths or other castes holding an equally high status, nor among the 
agricultural and higher artisan castes who come immediately below them in rank. It 
would be almost, if not quite, as difficult to find an Ojha among the pastoral castes, 
such as Ahir, Giijar, Ac, who rank immediately below the agricultural. The only 
castes then, who have retained their own sorcerers, and do not employ Brahmans for 
this purpose, are those which constitute the lowest stratum of the population, viz., the 
hunters, trappers, fishermen, scavengers, basket-makers, hide-skinners, &c, who are 
the least removed from primeval savagery and the farthest removed from the Hindu 

No worship can be paid to Kdli, the patron goddess of the Ojhas, without the 
shedding of blood. The animal now chiefly sacrificed by the low caste or aboriginal 
priest is the pig, that by the Ojha Brahman the goat. But the Tantrik sacrifice of the 
goat must not be confounded with the Vedic one alluded to above in para. 112. The 
ceremonies are totally different. There are, as is well known, two sets of mantras 
or sacred texts in Hinduism, one of which is called Vedic and the other Tantrik, 
and the latter only can be used in the goat sacrifice in honor of Kali. Moreover, the 
Vedic sacrifice can only be performed by Brahmans of the highest stamp (the Hotri 
and Ach&rya>, who would disdain to associate with an Ojha, and only in honor of the 
older divinities of Hinduism, amongst whom K&Ii and her consort, Shiva, had no place. 

The Kdlika Parana, which is the chief authority for the rites to be paid to K&li 
prescribes many kinds of animals besides the goat or pig, as fit to be sacrificed in her 
honor ; and the list is such as to show the savage or non-Aryan origin of Kili 
herself : " birds, tortoises, crocodiles, hogs, goats, buffaloes, guanas, porcupines, and 
the nine kinds of deer, yaks, black antelopes, crows, lions, fishes, the blood of 
one's body, and camels are the sacrificial animals," But the list of victims does 
not stop here. The same Pur&na breaks out into a rhapsody of delight on the 
merits of human blood :— " man, through my good fortune, thou hast appeared 
as a victim ; therefore I salute thee, though multiform and of the form of a victim. 

Digitized by 


(65 > 

Thou by gratifying Ohandik& destroyed all evils incident to the giver. Thou, 
a victim, who appearest as a saorifioe meet for the Vaishnavi, hast my saluta- 
tions. Victims were created by the self-born himself for sacrificial rites. I shall 
slaughter thee to-day, and slaughter at' a sacrifice is no murder/ 9 The practice 
of sacrificing men or children to K&li was once widely practised by the Ojha 
Brahmans in all parts of Hindustan and Bengal, and perhaps even now it has not 
entirely died out " Persons are not wanting," says Dr. Rajendra Lai Mitra, " who 
suspect that there are still nooks and corners in India where human victims are 
occasionally slaughtered for the gratification of the Devi; 1 ' and be adds that 
" there is scarcely a respeotable house in all Bengal the mistress of which has not 
at one time or other shed her own blood under the notion of satisfying the goddess 
by the operation" [Indo-Art/ans Vol. II., para. 111). The writer is here alluding to 
customs which prevail among Hindus of respectable castes ; but I have reason to 
think that the shedding of human blood his survived among the non-Hindu tribes 
also* In the course of the enquiries made by myself amongst the casteless tribes of 
Upper India, such as Kanjars and others, the hesitation with whioh answers were 
given respecting the kinds of victims offered to their tutelar goddess, Mari, Chand- 
rika, Kalika, Ac., all of whom are merely variations of the now Hinduized Kfili, 
leads me to suspeot that human sacrifices are still offered by the aboriginal priest 
in places where the act is not likely to be discovered. 

123. The Panda* is an inferior class of Brahman, whose special function con- 
sists in taking charge of temples and assisting visitors to present their offerings 
to the shrine. He might be simply defined as a temple* priest The word Pandd 
means science, and is the same as that from which Panda or schoolmaster and Pan- 
dit or learned men are derived. But the name Panda as applied to temple-priests 
has now become a misnomer; for all such men are totally illiterate— inferior in 
this respect, to priests of the Ojha caste, amongst whom literary ability may not un- 
. frequently be found, though here, too, it is decidedly on the wane. Temple-priests 
subsist for the most part on the offerings made to the idol at whose temple they preside. 
Such offerings may consist of money, grain, cloth, vegetables, fruits, live animals, 
such as the goat, the horse, and even the elephant. All of these are appropriated 
by the temple-priest, who in some cases, if his temple is much frequented, becomes 
very wealthy. The popularity of a temple depends, not on the degree of attention 
which the priest prays to it, but on its reputation for sanctity or antiquity, and on the 
rank of the god to whom it is dedicated. A Panda", who has amassed wealth through 
the offerings made at his temple, seldom remains there himself, but hires out some 
poorer Brahman to act as his proxy. If, however, he hears that some raja or other 
rich man is about to visit the temple, he takes goad care to be there himself and 
secure the largest. share of the liberal offering or fee that is expected. Almost every 
"Panda has a distinct circte of clients living at various distances from his own house or 
temple, and sometimes at a distance of one hundred miles or more. He endeavours to 
pay each of them a visit at intervals of one or two years, in expectation of the fee which 
clients so visited are accustomed to give, No Pan id is ever allowed to visit another 
Panda's clients. Nor may the client himself pay his devotions to any temple other 
than that at which his own PandA presides, unless it be to a temple dedicated to 
some other divinity. 

The Pand& is not able, however, tc appropriate all the offerings made at his 
temple to bis own individual use. He is expected out of these offerings to keep the 
temple in repair, in case no pious layman comes forward to relieve him of the bur- 
den. He is expected in any case to supply the clarified butter or oil with which in 
some temples an ever-burning lamp is fed, to provide the daily offerings of cooked 
food with which the hunger of the idol is supposed to be satisfied, to sweep out the 
interior of the temple, and to provide the bell which he rings at the stated hours of 


Digitized by 


( 66 ) 

Every temple in India, until it is deserted and oferings cease to be made to it, fe 
famished with an attendant priest. If the temple is attached to a private house 
and is not open to the general public, the purohit or family priest may discharge the 
necessary functions, or some Pandi may be appointed from without. In a public 
temple the Pandi first placed in charge of it by the founder acquires through long 
custom a prescriptive right to appropriate all the profits which he can make out of it, 
and to bequeath this right to his children and grand-children. 80 entirely is this right 
regarded as the private property of the priest who has acquired it, that he can use it 
as security for a loan, or sell it to any other Brahman, or to a GosMyen, or even U 
a layman ; and the layman would in this case put some Brahmen there as hk servant 
to receive the profits which the temple may bring in to him. 

The office of Pandi is not considered respectable by other Brahmans or by the 
upper castes of laymen. One reason of this is that the majority of the temples of 
which Pandas are in Charge are dedicated to Mahadev; and it is an old maxim 
amongst Hindus <rfthe upper castes, that offerings made to the lingam (the figure by 
which the presence of Mahadev is symbolized) will bring evil to any one who receives 
them, though they will bring good to the giver. It is said that when Daksha's altar 
was overthrown by Mahadev, the former uttered a curse that any offering paid to 
the lingam would prove an evil rather than a blessing to the priest who received it— 
a legend which corroborates the assertion made in the previous paragraph, that Ma- 
hadev was an aboriginal god, whom Brahmans for that reason at first declined to honor, 
but whom they were afterwards compelled to admit as the third member of the Triad 
in consequence of the ever-increasing absorption of the aboriginal tribes. In the 
Vedic hymns, where the Aryan side of Hinduism is specially represented, Daksha 
appears as one of the numerous forms of the Creator, while Mahadev or Shiva is never 
once mentioned. It is only by the upper castes of Hindus that the Pandi is not res- 
pected : for the lower wastes, who make up the great majority of the people, esteeifa 
him very highly as tJie mediator between themselves and their favourite divinity, Ma- 
hidev. Men whose caste is so low that they are not aUowed to go inside a templfe 
employ the Pandfc to place their offerings on the shrine. 

The most celebrated of the shrines in Upper India and elsewhere have distinct 
families of Pandfis attached to them, and the local groups made up of such families 
might truly be said to constitute separate castes ; for they marry only among them- 
selves and carefully exclude outsiders from participation in the privileges which they 
have secured. There are, for example, the Pandas of Gaya, who are called Qayawdls ; 
the priests of Mathura, who have taken the title of Chaube ; similar castes of Pandfis at 
the great temples in Benares, at Vindhyachal near Mirzapur, at Jwfiia Mukhi in the 
Kangra valley, at the temple of Kdli in Calcutta, at the temple of K&n&kshi in Assam, 

at the celebrated temple of Mah&dev in Golagokaran, and at many of the holy places 

of Ayodhya. 

One of the duties of the Pand4 is to sacrifice goats to Kali— a function which he 

shares with the Ojha Brahman whom we have already described. 

124. The Gangaputra or Ghatiya (for the two names are synonymous) might 
be called a river-priest. The former name means " son of the Ganges," and the latter 
" one who sits on the gh&t or bank." His special function consists in helping pilgrims 
to bathe in some sacred stream or tank during the appointed periodical festivals. 
His rank among Brahmans is about equal to that of Pand£; and their duties are very 
similar, except that the one presides at temples and the other at bathing-pl*oes« 
As the Pandi provides the oil and lamps required for the idol aad sometimes repairs 
the temple itself and assists the visitor to present his offering, so the Gangaputra 
repairs the steps leading down to the water, spreads a carpet or mat for the bathers, 
and takes charge of their clothes and shoes tilt they come out of the water. There are 
some Brahmans who combine the functions of river-priest and temple-priest in one* ; 

Digitized by 


( 67 ) 

such for example are the Chaubea of Mathura, the Gayawfils of Buddha Gaya, the 
Pr6gw61s of Allahabad, and others. 

The Gangaputra, like the PandA, is totally illiterate. The bather is supposed 
to have a manlra or sacred text recited over him by the priest as he descends into the 
water. Bat many of the priests cannot repoat even this. The pilgrim is generally 
satisfied, so long as he gets a dip into the holy water and pays a fee to the Brahman. 
By so doing be propitiates the water and washes away his sins. Biver-priests as a 
class are distinguished for their lazy and licentious mode of life ; but this dees not 
diminish their influence among the community at large. The pilgrim on arriving at 
•the sacred tank or river is at once pounced upon by some priest who has been watch- 
ing his approach, and is guided implicitly by what this man tells him to do. The 
stairs leading down to the water are parcelled out in certain lots, each of which is 
claimed by some particular priest as his hereditary property; and if one priest is found 
poaching in another's preserves, the encroachment may lead to riot and even to blood- 
shed. The property in a ghdt or bathing-stair can be given as a dowry, or sold, or 
mortgaged, or sub-divided between brothers, and in short treated like any other kind 
of private property. Sueh property is even recognized in the Government courts ; 
for both Pandas and Gangaputras will appeal to these courts for redress, when they 
ihink that their rights are being encroached upon by men of their own fraternity* 

It is specially important in the eyes of orthodox Hindus to bathe in the Ganges 
or some ether sacred river on days of eclipse ; and it is at the same time especially 
degrading to Brahmans to accept of fees or allies on days when the demon of dark- 
ness is abroad. But the Gangaputra has no scruples on this account, and this is an 
additional reason why he holds such a low rank among ether Brahmans. On the 
eclipse of the moon, the most lucky site, at whioh a person can bathe in the Ganges, 
is Benares? and on the eclipses of the sun, Euraksetra. The most appropriate kind 
of donation to be given on days of either eclipse is a cow ; but those who are unable 
to afford such a eostly offering give presents of grain, or cloth, or money. It is be- 
lieved that anything given on a day of eclipse will be returned a hundredfold to the 
giver. Hence Hindus esteem themselves fortunate in having been provided with a 
priest of the Brahman order, who will accept of donations on such inauspicious days. 

Another function for which the services of the Gangaputra are often employed 
consists in the assistance rendered to ignorant and illiterate men in making the 
annual offering to the souls of ancestors. An offering of water and of the cake 
or ball of rice called pinda is paid once a year in the month of Ku&r (partly Septem- 
-ber and partly October) : this offering is repeated for fifteen days continuously and is 
♦called srtidiha, while the days set apart for paying the offering are called pitri-puksh*. 
Hie priest leads his clients into the water, shows them what to do, and utters the appro- 
priate text if he knows it, or invents one if he does not. Men who are educated do 
not seek his assistance, butihey do not withhold the customary fee for the privilege 
of using his gh&t or bathing-stair. 

125. The Joshi is a caste of Brahman who professes the art of telling fortunes. 
The name is merely a contraction of Jyotishi, " astrologer.'* His art, though somewhat 
akin to that of the astrologer, is not by any means the same; but writers have been apt 
to confound them. For instance, in the Nerth-West Provinces Gazetteer, Vol. VIII, Part 
II page 1, it is said: "Joshi is a class of Brahmans who follow astrology as a pro- 
fession and earn a subsistence by casting nativities. 9 ' This is true not of the Joshi, but 
of the Jyotishi ; for the Joshi never casts nativities and is not really an astrologer. 
Again, in the same Gazetteer, Vol, V, page 583, it is said that "the trade of the Jyotishi 
is fortune-telling or astrology. The planet Sanichar or Saturday is their favourite 
deity. Ac." These remarks apply to the Joshi, but not to the Jyotishi ; for the Jyotishi 
is not under the patronage of the planet Saturn, whereas the Joshi is. The difference 
between the two men is this : the Jyotishi tells fortunes by the stars, while the Joshi 

Digitized by 


( 68 ) 

does so by the lines and other marks on the palms of the hands, on the face, and on 
the body generally. The nearest term, then, to express the Joshi's art is palmistry, 
while the exact term expressive of the art of the Jyotishi is astrology. 

Another mistake has sometimes been made in supposing that the Joshi is a mere 
migratory impostor of the gypsy stamp, having no historical foundation for the folia- 
tion which he affbpts to practise. The supposed analogy between the Joshi and the 
gypsy is entirely groundless* The art of palmistry is a very old one in India and 
has a literature of its own. The caste, too, has as good a pedigree as any other caste 
in India. Nor is the art of palmistry more absurd than that of astrology, which is 
generally practised by high-caste Brahmans. The old name for palmistry was 8a- 
mudrika, and in ancient times one who specially devoted himself to its study was 
called Samudriki. It was considered to be a collateral branch of the single science of 
Jyotish, and this is the reason why the caste is now called by the name of Joshi. 

The Joshi of modern times has become totally illiterate ; yet he is generally to 
be seen with a manual of palmistry in his hand of which he knows nothing. The 
highest castes of Hindus refuse even at times to consider him a Brahman ; but he 
regards himself as one, and is so regarded by about eighty per cent, of the population. 
Moreover, he wears the sacred thread, and has worn it from a remote ancestry : and be 
is by no means the only Brahman with whom other Brahmans refuse to associate. 
The fee which he receives for delivering bis oracle (the wording of which, like those of 
Delphi, is always studiously ambiguous) is generally a supply of grain sufficient for one 
day's consumption. Bnt if one of bis clients happens to fall in with some stroke of 
good hick, the Joshi at once pays him a visit and attempts to convince him that it has 
occurred in fulfilment of his prediction, and demands a special fee for having proved 
such a wise and auspioious prophet. Each Joshi has a select circle of constituents 1 , 
who live in villages surrounding his own at a distanoe of about ten or twelve miles, and 
no other Joshi is allowed to visit them. In the hill districts of Knmaun, where some 
of the old Hindu customs have retained something of their pristine vigour, the Joshi 
is more respected than be is in the plains ; and in Knmaun he is not so illiterate. In 
those districts many of the clerks in the Government offices are of the Joshi caste, 
and Pandit Mathura Dat, late head-master of the anglo- vernacular school of JalaiH*, 
might be quoted as an example of a Joshi who has done credit to his fraternity. The 
name Bhanreriya or "gabbler" has been sometimes given to men of this caste, on 
account of the fluent readiness with which they read the- fate of a person after examining 
his hands and face* 

The art of the Joshi has (as was stated) a certain connection with astrology, and, 
so far as I can learn, the connection is as follows :— In the Hindu system of astronomy 
there are said to be nine planets, viz., the five regular planets, the sun and moon, and 
the two demons of the eclipse. The collective name for these nine planets is Nava- 
graha {nata meaning "nine" and gratia "planet"). These planets, like the stars of the 
lunar zodiac, are believed to exercise an extraordinary influence upon human destiny. 
Three are said to be auspicious and are called Subhagraha; three others are said to bo 
less auspicious and are called Pipagraha; while the three last are said to be cru*l 
or malignant and are called Kruragraha. The scheme stands thus: — 

( Guruwir ... „ Thursday (Jupiter). 

Subhagraha ... I Somwsr ... ... Monday (the Moon). 

CSukrawar ... M . Friday (Venus). 

(Bhaumwar ... .„ Tuesday (Mars). 

Pspagraha ... < Budhawar ^. „, Wednesday (Mercury). 

C Adityawir (Itwar) ... Sunday (the 8un). 

r Shanaischar (Sanicbar) ••• Saturday. 
Eruragraba — \ Kahu > _ _ . _ 

C Ketu > *** ••• Beacons ol the eclipse. 

It is with the last three planets alone that the Joshi is in league. The offerings 
made to these malignant powers and transmitted to them through their appointed 
priest, the Joshi, consist of oil, the black pulse called urd f pieces of iron, black cloth, Ac. 

Digitized by 


( 69 ) 

The colour black is the appropriate emblem of those deities of darkness, and oil for 
relieving the darkness is the appropriate offering. It is customary for the Joshi to 
receive such gifts on the Saturday. 

126. The last and lowest caste of Brahman is the funeral-priest, who, in conse- 
quence of the aversion with which he is regarded by all classes of the community, 
is contemptuously termed Maha-Brahman or Maha* Patra, " the great priest or the great 
Tessel." He is sometimes called the Karataka or Vulture Btahman, because priests 
of this caste flock like Vultures round the carcasses of the dead. Nevertheless, the< 
function which the Mah&-Brahman performs, as will be seen below*, is a very impor- 
tant one in the eyes of Hindus ; and the very highest castes, though they consider it a 
pollution to touch him, cannot dispense with his services. Such are the strange 
inconsistencies of the supernatural creeds : the Mahd-Bfahman is indispensable for 
discharging the pious offices due to the dead) and yet he is loathed for the very reason 
that he performs them. 

Amongst Hindus, as amongst all other people whose religious beliefs are in the 
savage or barbarous stage, the soul of the dead is supposed to suffer from hunger and 
thirst, and to need the same conveniences- that it enjoyed in the body which it lately 
occupied. To this sentiment, for example, must be ascribed the atrocious rite of $ati 
or the burning of the widow alive upon the pyre of her husband, so that she may ac- 
gompany him to the world beyond. Thi^rite has long been suppressed by the British* 
Government i but a series of less mischievous rites has survived of whicb*alt was only 
one link in the chain* On the day after a person has died the survivors hang an 
earthen vessel called ghant from a tree — a pipal tree, if one can be found. This vessel 
is replenished every morning and evening with water, while a small lighted lamp, 
intended to give light to the departed soul, is placed on the top of it. A small hole 
fc bored at the bottom, so that the water may trickle out to appease the thirst of 
the dead. The vessel continues to hang and to.disoharge water for ten, twelve, fifteen 
or thirty days, according to the rank of the caste to which the deceased belonged. But 
the operation would be of no avail, unless the Mahd-Brahman himself suspended and con- 
secrated the vessel. All this- while every member of the deceased's family is considered 
unclean and must bathe every morning at some particular hour and as near as possible to 
Ae place where the vessel is hanging. The Mah4-Brihman is always present at such, 
times, and, after placing a bunch of dry kusha grass on the ground, causes every mournec 
or bather to throw upon it an offering of water and a handful of black til seed for the 
uefreshment of the departed soul. The name of this ceremony is called tilanjuli, be- 
cause the til and water are thrown from the joined palms of the hand (anjuli). For 
assisting afethis daily ceremony the Mahi-Br&hman receives every day enough cooked or 
uncooked grain to keep him for the day. After the ten, twelve, fifteen, or thirty day* 
have expired, the Maha-Brihman goes to the tree where the vessel is hanging, and, after 
breaking it in the presence of the assembled mourners, is presented by them, with every 
kind of thing that the departed soul is likely to require in the next life, such as tobacco, 
grain, clotKes, carpets, pillows, bedsteads, shoes, walking-stick, the hukka or native 
pipe, cooking utensils, Ac. If the deceased's survivors can bear the cost, they provide 
a new batch of all such articles, besides giring«(as all must do, whether rich or poor) the 
old clothes, the old pillows, the bedstead, the cow or buffalo, the plough-cattle, the 
palanquin, &c, which the deceased was specially accustomed to use, If a R4ja or rich 
man dies, the MahkBr&hman receives even his horse and elephant. The meaqing of 
all this is that the departed soul requires, or is believed to require, after death, everything 
that he used during life, and that the MahA-Brahman is the medium through whom they 
are supposed to reachhim. In every country but India such thingsure buried in the same* 
tomb with the body ; and it is chiefly througbthe contents found in the interiors of tombs, 
that the arts and inventions known to the ancestors of mankind have been discovered by . 
archaeologists. But in India no tombs are erected. The bodies of the dead are burnt,, 
and the ashes are thrown on the rivers; ichife the Mahd-Brahman acts the part ot 


Digitized by 


( 70 ) 

the tomb, being himself the living tomb which receives the gifts intended for the 
happiness of the dead. A better example could scarcely be quoted of the extent to 
which Brahmans have traded on the instincts natural to the human mind and turned 
them entirely to their own advantage. 

On the twelfth day immediately following the cremation of the corpse a horde of 
Mahd-Brahmans is summoned to eat a banquet, and on the thirteenth (Jay other Brah- 
mans belonging to different classes are invited for the same purpose. Thus the 
feast to the dead, which in every other country is eaten by the kinsmen of the 
departed, is in India eaten by Brahmans* 

THe Mahi-Brahman as a class is totally illiterate : he can seldom even repeat 
correctly the texts which are supposed to be uttered during the different stages of 
the ceremonial. 

127. This completes the long list of priestly functions exercised by the modern 
Brahman. It must now be shown to what extent these several functions have given 
rise to corresponding castes or sub-castes within the Brahmanical aggregate. 

Among the nine different functions described under headings 1 and 2 in para. Ill, 
there is none that now represents a caste or sub-caste in the strict sense of the term. 
But as Dikshit, Pathak, and P&nde are and have long been the hereditary titles of 
corresponding classes, and as each of these classes still prefers to marry within its own 
ranks if suitable connections can be found, it is more than probable that each of them 
was once actually a caste or sub-caste — that is, a body of men exercising a function 
peculiar to itself and allowing none of its members, either male or female, to marry 
ontside its own fraternity. Ach&rya is an hereditary title in Bengal. Jyotishi must 
once have been a distinct caste in Upper India ; for otherwise the name Joshi, by 
which the kindred caste of palmisters is called, could not have come into existenoe. 
Among the names descriptive of the function of sacrifice there are two, viz., Bajpay 
and Jijhotiya, which are almost synonyms of Hotri. The former is the hereditary 
title of a certain Brahmanical class which like Dikshit, P&thak, Pande, &c, must 
once have represented a separate sub-caste. Jijhotiya is the name of a distinct Brah- 
manical caste, with which no other caste will intermarry, dwelling chiefly in the 
plains between the Ganges and the hills of Central India. 

Among the five different functions named under heading 3 in the same paragraph, 
there is none that does not represent a corresponding caste in the strictest sense of 
the term. The Ojhas, the Pand&s, the Gangaputras, the Joshis, and the Mah&-Brah- 
mans, are all entirely distinct units — distinct not only from each other, but from all 
the other classes of Brahmans named under headings 1 and 2. In the case, then, 
of these five classes of priests, viz., the sorcerer, the temple-priest, the river-priest, the 
palmister, and the funeral priest, the law maintained throughout this paper, by which 
diversity of function gives rise to diversity of caste, is seen in its most perfect opera- 
tion. Probably, too, its operation would have been not less complete in the case of 
the other nine functions also, if the field within which they were practised had not 
been pre-occupied by the ancient system of sub-divisions into gotras, orders, clans, 
and branches, to which allusion was made in paragraph 110. This system had beefr 
formed and fully established before the modern developments of Brahmanism had 
come into existence. It was based, as we have shown, not upon diversity of priestly 
functions (for in the Yjedic age such diversities did not exist), but upon supposed 
diversity of descent from the seven great Rishis or sages. During the long struggle 
with Buddhism which followed the close of the Vedio revelation, Brahmans were more 
than ever careful to preserve every time-honored tradition that existed in regard to the 
organization of their caste ; and hence this system not only survived the conflict 
with the rival creed, but has lasted to the present day. Thus when the new develop- 
ments of Brahmanism had taken their present shape, the old system of sub-divisions 
into gotraa and clans and branches was too strongly established to admitofnew castes, 
based upon the new diversities of function, being permanently formed. 

Digitized by 


( 71 ) 

128. The statistics given in the census report regarding the Brahmanical sab- 
castes are very incomplete. No separate mention has been made of the Ojha, the ' 
Pandi, or the Gangaputra, all of which are (as we have stated) distinct social units. 
The only sub-castes of whom separate statistics have been given are the Joshi, the 
Mah&- Brahman, and the Oharaj. The last two, however, are really one and have been 
shown as such in the accompanying statement ( group VI), Charaj is merely a corrup- 
tion of Ach&rya, a high sounding title assumed by funeral-priests, not less pretentions 
than that of Mabfi-Brahman or Maha-Patra. 

129. The reader must have been struck with the contrast between the complex 
developments of Brahmanism which are in force at the present day and the single 
function of sacrifice which marked its earlier career. But there is one common 
element which runs through all these various phases of the Brahmanical office. 
Except in the case of the schoolmaster, viz. y the Pathak and Pande, who were selected 
from among the Brahman caste merely because Brahmans alone were sufficiently edu«- 
rated and sufficiently sacred to be trusted with the training of the young, every variety 
of Brahman that I have described in this paper acts as mediator between man and the 
invisible world ; and to this extent the caste is on a par with the priests of other 
creeds. Thus in Vedic times the Hotri was the mediator between the sacrificer and 
the Devas or gods who personified the elements. In modern times the Bidua is the 
medium between the founder of a temple and the deity whom he draws into the idol ; 
the Dikshit between the boy whom he initiates and the gods to whom he dedicates him ; 
the Jyotishi between the man who consults him and the stars whose influence he inter- 
prets ; the Pauranik between the man who listens to his recitals and the spirit presid- 
ing over the book from which he reads ; the Ojha between the man who is possessed 
with a devil and the death-goddess t hrough whom the devil is to be expelled ; the Panda* 
between the man who brings an offering to the shrine and the deity at whose shrine 
it is made ; the Gangaputra between the man who bathes hi the sacred pool and the 
spirit presiding over the water ; the Joshi between the terror-stricken suppliant and 
the demons of the eclipse ; the Maha* -Brahman between the living and the souls of 
the dead ; while the Purohit can mediate for bis employer in almost any capacity, even 
to the extent of bathing or fasting for him. Thus the Brahman is a true priest, in 
whatever direction, except that of schoolmaster, his services have been employed : and 
even when he acts as schoolmaster, he does not resign his position as priest or messen- 
ger of men to the unseen world. That this is the light in which the laymen of all 
classes regard him is clear from the fact that, in whatever capacity hfe may be employ- 
ed at the present day, he is still invariably called jackak or 'sacrificer, while the man 
for whom he officiates is called yajaman, or the man who pays for the sacrifice. Thus 
the astrologer, the sorcerer, the river-priest, the funeral-priest, &c, &c, are at bottom 
priests of the sacrifice-s-messengers through whom the oblations of man are transmitted 
to the abodes of gods and departed souls. 

130. There is one rather peculiar rite by means of which the Brahman is 
used by all classes of Hindus as messenger between gods and men, aud this is by • 
making him eat — a ceremony, which so far as I know, is confined to the Indian 
priesthood and is altogether unknown in other countries. In the ancient or Vedic 
age the element on which the flesh and butter of the saorifice were thrown was fire. 
But the Smritis, or books of canon law (of which Brahmans were, of course, the authors), 
are never weary of telling us that the Brahman himself is " the flame of the sacrificial 
mouth," and that if the leavings, that is, the solid parts, of the sacrifice go to any but 
Brahmans, " the offering is not made to gods (Devas), but to devils (A suras)." Thus 
the Brahman is a consuming fire ; and provided he eats at any one's expense but his 
own, he is one of the chief means by which at the present day men of other castes 
transmit their offerings to the higher powers. So whenever a man wishes to perform 
some expiatory rite, or to invoke the divine favour upon some new undertaking, or to' 
signalize some important event in his own life, such as a birth, a marriage, or a 

Digitized by 



( 7* r 

fiinerar, he invariably invites a horde of Brahmans to a banquet. It is not at all essen- 
tial to the efficacy of suoh banquets that the Brahmans who are fed should be men of 
cultivated minds or versed in sacred science of in the practice of priestly offices* 
Whatever his training or mode of life may be, he is by birth an impersonation of the 
deity, a messenger between gods and . men. " Whether literate or illiterate/' says. 
Krishna in the Bhagavad Qiu*, " a Brahman is my own body." To feed a Brahman, 
therefore, is to feed Krishna or the- divine being himself. The object of all sacrifice 
in the Vedic age was to feed the gods. Bnt as the slaying of animals to any deity 
except Kali is now practically extinct, the rite of feeding the gods by feeding Brah- 
mans has succeeded to its place. 

131. After all that has been said about the functions which in ancient as well 
as in recent times have beea the peculiarities of the Brahman caste, it may cause- 
some surprise to the reader to be told that the most hon&rable state in which a Brah- 
man can exist is to live as much, as possible for himself and exercise no. functions at 
all for the outside community. Such, however, is the ideal set forth in Mann's Code,, 
and such is the ideal iecongni»ed at the present day. 

There axse several- considerations which commend this view of a Brahman's sta- 
tus to the minds of Hindus. As there is no caste above that of Brahman which can- 
perform religious rights on his behalf, such as he himself is able to perform for others,, 
any time that may be spent on. ministering to the wants of others represents a loss* 
of grace and opportunity to himself, and he suffers in holiness and purity accordingly. 
Again, there is nothing so degrading to a service* Devotion to the good of: 
others is a sentiment entirely alien, to the spirit of bis creed ;. for it is a. fundamental 
maxim of the Brahmanical codes to believe that all other races were created to serve 
himself. But whenever a Brahman undertakes to discharge some priestly function 
for another, he becomes for the time being the servant of the man who employs him. Ifo* 
Parohit, for example, is the servant of his master for life, and this is one of the chief 
reasons why the office of hereditary family-priest is considered so degrading.. Again, 
it is impossible for any one rtho desires to propitiate some deity to do this without 
presenting an offering in oaskor kind ; and this offering is appropriated by the Brah- 
man, who receives it on, behalf of the deity, being himself an impersonation of the 
divine being. But, in spite of all theories to the contrary, the constant acceptation of 
gifts has a mercenary look, and the receiver loses in dignity as much, as the giver 
gains in grace. The highest status, then, in which, a Brahman can live is to stand en- 
tirely aloof from, the outside world, a priest only to himself and family, to lead, 
the model life made up of the daily routine of offerings and ablutions prescribed by 
the rules of his order, and, when his sons are grown up and married, to retire into a. 
hermitage and meditate till the day of his death on the mystery, of Brahma, the- 
Supreme Spirit. 

There are but few Brahmans, however, who can dispense with the emoluments of 
their calling : and as there was a never-ceasing demand among the general commur- 
nity for religious direction of all kinds, which only a Brahman could give, the author 
of Mann's Code was generous enough to lay down the rule that there are six kcrmcu 
or works in which the Brahmanical caste may lawfully engage, while to every othec* 
twice-born caste there are only three :.— 

(1) To study the Yedas ^pathan, 

(2) To teach the Vedna ; pdthan. 

(3) To offer sacrifice ; yo/an. 

(4) To help others to sacrifice ; ydjaiu 

(5) To give alms ydana. 

(6) To receive alms or fees ypratigrah* 

The three duties which any twice-born caste in common with the Brahman may per- 
ioral are the first, third« and fifth.. The three which only a Brahman, and no other 

Digitized by 


( 78 ) 

caste, may perform are the second, fourth, and sixth ; and he is not allowed to per- 
form any of these, unless he is in want of the necessaries of life. Mr. Sherring makes 
the curious mistake of saying that " only those Brahmans who perform all these six 
duties are reckoned perfectly orthodox. Some perform three of them, viz., the first, 
third, and fifth, and omit the other three ; yet they suffer in rank in consequence" 
(Hindu Tribe$ and Castes, Vol. L, page 10). Such a statement is directly opposed not 
only to existing facts but to Manu's Code (X, 103), which runs thus : u From teaching 
the Veda, from officiating at sacrifices, or from taking presents, though these things 
are generally disapproved, no sin is committed by priests in distress ; for they are a* 
pure as fire or water." 

132. The reader need scarcely be reminded that in the aocount which we have 
given of the several functions open to the modern Brahman we have merely consider- 
ed him as a priest, and taken no account of the fact that many members of the caste 
are in practice mere laymen, who discharge no priestly functions whatever. Out of 
the total number of Brahmans recorded in the census of 1881, viz., 4,690,850, the 
number of males, that is, of potential priests or Levites, is given as 2,443,040. 
Turning to the details of occupation given in the same census, we find that only 
81,318 persons are set down as " Hindu priests." If to these we add 509 more who 
in another place are shown as astrologers under the rather misleading title of " scien- 
tific persons," we get a total of 81,827 (see pages 105 and 106 of Preliminary Disserta- 
tion). According to this the proportion of actual priests to the total number of poten- 
tial ones would be less than 4 per cent., and this is obviously too small to be correct. 
So far as I can learn from persons likely to be well informed on the subject, Brah- 
mans of the present day might be sub-divided as follows : — 

(a) Those who live exclusively by the practice of religious functions ; about 
fifteen per cent. 

(&.) Those who live partially by such functions, but follow various secular call- 
ings in addition ; about twenty-five per cent. 

(t.) Those who, without performing any of the priestly offices implied in the 
above, might yet be asked to sit down to a banquet and eat at another's 
expense ; (for even this must be accounted a priestly function in India) : 
these make up some twenty per cent more. 

(d.) Those who perform no priestly office whatever, not even that of eating : 
these make up the remaining forty per cent. 

133. According to Manu's Code the state of life most becoming to a Brahman 
was, as we have shown, to be a man of independent means and do nothing but live 
for himself and for his own soul. But if he was poor, he might earn a subsistence by 
" assisting to sacrifice, teaching the Vedas, and receiving gifts from a pure-handed 
giver" (X, 76). But supposing even these did not suffice, " he might live by the duty 
of a soldier, for that is next in rank" (X, 81) ; or if this failed, " he might subsist as 
a mercantile man, applying himself to tillage or attendance on cattle " (X, 82). 
But he might never plough the field with his own hand ; for the " iron-mouthed 
pieces of wood not only wound the earth, but the creatures dwelling in it" (X, 84). 
Nor might he ever take up secular work for hire as the servant of another, " for 
this is svtavritti or dog-living " (IV, 6). 

With the exception of the last, each of the above rules is observed for the most 
part at the present day. There are certain classes of Brahmans, the Fande for exam- 
ple, which are distinguished for their military propensities ; and the same may be said 
of almost all the Brahmans, of whatever class, living in the Baiswara districts of Oudh. 
Commerce, too, is a favourite occupation of the caste : and there are few, if any, forms 
of trade in which a Brahman will not engage. Many follow the occupation of milk- 
man and cattle-grazer ; and as the cow is a sacred animal, this is not uncongenial to 
the instincts of the caste. If Brahmans take to agriculture, as many do, in the capa- 
city of either landlord or tenant they will in some few districts, chiefly in those to the 


Digitized by 


( 74 ) 

south of the Ganges, plough with their own hands ; but elsewhere they employ Cha- 
m&rs, Koris, or Ah(rs to do this for them ; for they still cherish the tradition that 
ploughing is forbidden to the caste. As to the " dog-living" or working for hire, all 
aeruple against such a practice appears to have completely died out. Brahmans may 
now be found in any and every kind of occupation for which wages are paid, except 
those which would entail ceremonial pollution and consequent loss of caste, such as 
that of sweeper or washerman. They will act as water-carriers, cooks, cart-drivers, 
night-watchmen, field-watchmen, messengers, policemen, public singers, dancers, 
wrestlers, Ac Latterly, too, they have taken largely to thieving and other forms of 
crime. Grasping and lazy by long hereditary instinct, and not being so liberally sup- 
ported by the outside community as they consider that they ought to be, they have 
begun to resort to force; and in Oudh at least they are now one of the chief criminal 
classes, ranking in this capacity scarcely, if at all, below Pfisis. 

In spite of all this, every Brahman, even the lowest, is still called Mab&raj, or 
great king. Every other caste still looks to him for his blessing as he passes. The 
coarsest features, the most abject ignorance, and the most humble occupation do not 
detract from the fact that in the eyes of the Hindu community he impersonates Brah- 
ma, the Supreme Being, and is entitled as such to the homage of mankind. A 
wretched Brahman once complained to me of the hardness of his lot, that " the God 
had descended in him in vain.' 9 

134. There is one class or rather caste of secularized Brahmans whose here- 
ditary occupation is of such a peculiar nature as to deserve a passing notice. These 
are the Sanauriyas of Bundelkhand, a caste which claims affinity with the great Sanad- 
hya stock whose chief habitat is the plain between the Ganges and the Jumna. 
The hereditary function, if it can be called so, of this caste is thieving, but only by 
daylight. They make no secret of the fact that this is the main industry by which they 
live, and they even defend it on quasi-religious grounds. They quote a legend which 
shows that this was the lot assigned to them by Ram Chandra of Ayudhya, the'great- 
est of the incarnations of Vishnu. There is a Hindu proverb to the effect that a theft 
by a Brahman is a gain rather than a loss to the person robbed :— 

" Girpare Gang a ; Churai Khai Bamhan" 

That is, property stolen and eaten by a Brahman should be no more regretted 
than property which has fallen into the Ganges. To eat at ' another's expense is, 
as we have shown, a priestly function in India. To give what is eaten voluntarily is 
of course more meritorious than to have it taken from you by stealth ; but to be rob- 
bed by one who impersonates the deity is much better than to be robbed by anyone 
else. The Sanauriya throws an odour of sanctity over the act by the rule of robbing 
only in daylight— a rule so jealously observed by the caste that any infringement of 
it would entail dissmissal from the fraternity. JNighfc is associated in the minds of men 
with the commission of evil deeds, and the Sanauriya by performing the daily argh or 
oblation of water to the Sun-god and by repeating the Gayatri as other Brahmans do, 
secures, as he thinks, the countenance of the Sun-god to the predatory life which he is 

The goddess who is believed to preside over the peculiar craft of this caste is Kali- 
ka. In order to escape detection they have invented a thieves' Latin and a gesture 
language, with which, however, the police authorities are now beginning to be familiar. 
They carry their thievish expeditions sometimes as far as to Malabar, Bombay, or the 
Madras coast, leaving their wives to till the fields in their absence. The person by 
whom the theft is actually committed is generally a boy under twelve or fourteen years 
of age ; and such boys are taken regularly into training by the men, who teach them 
the sleight-of-hand necessary to the difficult art of stealing in broad daylight, and 
organize all their movements on march. 

The Sanauriya is a strict Brahman. He renews his sacred thread annually, as 
other Brahmans do, in the ceremony called Hakshabandan ; he is a total abstainer 

Digitized by 


(75 ) 

from flesh and wine ; never destroys life in any form ; observes the same ceremonies 
connected with births, marriages, and deaths that other Brahmans do ; worships the 
cow and the serpent, and when he is sick employs none but Brahman sorcerers to expel 
the evil spirit from his body. Whenever, he meets a man notrof the Brahman caste, 
he pronounces the usual arirbad or blessing and receives back the usual salutation of 
Mah&raj, or great king, athough he is known by every one to be a professional thief. 

1?5. The Brahman caste has with that of Chattris become one of the last 
strongholds of Aryan ethnology, so far as this theory of race has been applied to the 
Indian people. As the genesis and classification of caste which this paper is attempt- 
ing to unfold are based on the hypothesis of the unity of the Indian race, I am com- 
pelled, before dismissing the subject of Brahmans, to go briefly into the question as to 
whether any grounds exist for supposing them to be of Aryan as distinct from indi- 
genous or Indian blood. The question has already been discussed in regard to Chattris 
in para. 37 ; but as the same line of argument does not apply to Brahmans, their 
case requires to be separately dealt with. 

Originally, when Aryanism first came into vogue, all natives of India, of whatever 
caste, who spoke any form of neo-Sanskrit or Hindi, wore for this reason declared to 
be Aryans ; and the phrase l( Aryan brother " has passed into a proverb. No one 
would think of denying that Sanskrit and neo-Sanskrit are Aryan tongues ; and if the 
statement made by Professor Max Miiller is correct, that " the evidence of language 
is irrefragable, and is the only evidence worth listening to with regard to ante-histo- 
rical periods" (Ancient Santhrit Literature p. 13, odit. 1859), there is nothing further 
to be said. But it has become the fashion of late years, amongst writers of the same 
school, to say that only the upper castes of Hindus are Aryans, all the rest being 
either aboriginal or mixed. A moment's reflection will show thst the linguistic theory 
as thus applied refutes itself. The very lowest of Indian oastes, such as Cham&rs, 
P&sis, Bhars, Arakhs, &c, not to mention tribes still lower in the scale, suoh as Doms, 
Kanjars, Haburas, Ac, speak Hindi quite as much as Brahmans do, and from time 
immemorial have known of no other language. If, then, these oastes and tribes are not 
Aryan at all, and the test of language proves to be worthless in their case, it must 
be equally worthless in the case of the Brahman also* ' 

Convinced, however, that Brahmans at least, whatever the other castes may be, 
are of Aryan blood, many writers have learnt to speak of them as having fairer com- 
plexions and finer features than ordinary Indians, and find in this a further proof of 
their Aryan descent Mr. Sherring for example, speaking of the Brahmans of Benares, 
writes thus : — " Light of complexion, his forehead ample, his countenance of striking 
significance, his lips thin, his mouth expressive, his carriage noble and almost sublime, 
the true Brahman * * * is a wonderful specimen of humanity walking on God's 
earth." We can only meet this statement with a simple denial of the fact. A walk 
through the class-rooms of the Benares Sanskrit College, in which some four hundred 
students, all of the Brahman caste and hailing from all parts of India, south as well 
as north, are assembled, would convince any one who used his eyes that the great 
majority of Brahmans are not of lighter complexion or of finer and better-bred features 
than any other caste. The expression of the face may be more intelligent than that 
of the labourers working in the roads : but expression is the result of culture, and this 
is quite a distinct thing from diversity of physical type. A stranger visiting India 
for the fitst time, and walking through the Benares class-rooms, would never dream 
of supposing that the students seated before him were distinct in race and blood from 
the scavengers who swept the roads. But a man whp has lived long enough in this 
country and seen more provinces than one might discern that some difference of 
feature and general appearance exists among the different students, and that this 
difference depends on the nationality to which they belong. He would observe, for 
example, that a Bengali Brahman looks very like other Bengalis, a Hindustani like 
other Hindustanis, a Mahratti like other Mahrattis, and so on, which proves that the 

Digitized by 


( 76 ) 

Brahmans of any given nationality are not of different blood from the rest of their 
fellow-countrymen. It is not denied that rather fair complexions, recalling the 
Aryan type, do occasionally show themselves amongst men of the Brahman caste. 
Bat similar instances of atavism occur among the lowest castes also, and occur much 
less frequently among Brahmans than among some of the trading castes. Aryan 
blood has undoubtedly filtrated to some extent through all classes of the people. If 
Brahmans have rather a larger share of this than Cham&rs, the difference is not so great 
as to entitle us to speak of the two castes as belonging to different races* The truth, 
then, appears to be that the Aryans, who entered the river-basins of the Indus and the 
Ganges vid Cabul and Kashmir, became absorbed after two or three centuries in the 
pre-existing population, leaving, however, as evidence of their immigration some slight 
modification of the features and complexion of the native race, a language which has 
superseded most of the indigenous languages of India (as Latin or neo- Latin ha& 
almost ousted the Basque languages of Spain), and a religion which a few centuries 
later became profoundly modified, or rather completely transformed, by thorites, 
customs, and beliefs of the aboriginal savage. 

Some writers, again, who admit that Brahmans in certain parte are a dark-com- 
plexioned race, ascribe this fact to the effect of a long residence under a tropical sun r 
and believe them to be Aryans notwithstanding. Professor Max M tiller for example,, 
in discussing the physiognomy of Brahmans, observes: — "Time, hewever, has worked 
many changes ; and there are at present Brahmans, particularly in the south of India,, 
as black as Pariahs " {Chips from a German Worship, Vol. II., page 323). But the 
argument has the fatal defect of proving too much. If physical appearance is a mere 
matter of time and climate, and if through these causes Brahmans have become as 
dark as the ordinary native, we have nothing but the worthless evidence of language 
to show that they are Aryans at all. Moreover, the argument is opposed to facts. 
There are Jews in the south of India who have lived there for more than a thousand 
years, and are still as fair as the Jews of Palestine : these are called the white Jews. 
In the same part there is another community of Jews, holding no intercourse with 
the preceding, who are as dark as the darkest natives and are called the black Jews. 
The latter have taken wives from the native tribes, while the former have not. The 
Aryans who settled iu the plains]of India followed the same course as that adopted by the 
black Jews, and consequently their blood has been lost in that of the native population. 
On the other hand, the Aryans who remained in Kashmir and never entered the 
plains of India at all have retained their Aryan features and complexion to this day. 

No one can now be called a Brahman (as every is one aware), unless he can satisfy 
his neighbours that his pareuts on both sides were like himself of Brahman parentage. 
This is what is meant by caste, so far as birth is concerned. Assuming that the priests 
of the earliest or Vedic age were Aryans, writers have drawn the conclusion that the 
modern Brahmans, who by the rules of oaste are their hereditary descendants, must 
be Aryans also. But the restrictions of marriage which are now imposed by the 
rules of caste did not begin to exist until at least a thousand years after the Aryans 
had come into the country, and by this time the Aryan blood had been absorbed 
beyond recovery into the indigenous. It was not till the time of Manu, that is, 
about 200 B. C. or later, that the caste rules in regard to marriage were coming into 
force. Even then, as his own writings show, they were not universally accepted by 
Brahmans themselves : for he waxes very wroth with certain Brahmans of his own day 
who persisted in the habit of taking Sudra or low-caste women as their first wives, and 
dooms them in consequence to the most terrible penalties in the next life (III, 17). It 
is clear, then, that prior to his time, that is, ever since the Aryan invader had set foot 
on Indian soil, which must have been more than a thousand years before his Code was 
compiled, a Brahman or professional priest (for the Brahman caste did not then exist), 
could marry any woman that he liked. The hymns of the Big Veda Sanhita are said 
to be "the earliest collection of Aryan poetry;" yet some even of these were composed 
by a man of " Dasya " or aborignal descent (Max Muller's Ancient Sanskrit Literature^ 

Digitized by 


( IT ) 

page 58). The authors of such hyms were, in the Vedic age, called by the name of 
Brahman, because a hymn was called brakma. But if the author of a hymn oould be 
the son of an aboriginal mother, the word Brahman must have meant something very 
different in that early age from what it means now. The meaning which Manu laboured 
to fix upon it, or rather the rule which he laboured to establish as binding upon priests, 
was that no one could be considered a real Brahman, uulesfrhis-mother was of that class 
as well as his father. But as Brahn&ans of this perfect stamp were not always to be 
found, he was kind enough, to tell his countrymen that, in selecting a priest? for making: 
offerings to the gods* they should not enquire too closely into a Brahman's parent- 
age :— " For an oblation to* the gods, let not the man^ who knows what is law scrupu- 
lously enquire into the parentage of a Brabman ; but for an oblation, prepared for an- 
cestors, let him examine it with strict care " (III, 149). It is- quite clear, then, that 
even i» Menu's time, and a> fortiori an the centuries preceding it, the- gods were accas- 
tomed to be fed through any priest who was versed in- the intricacies* of the sacrificial 
art, and that the^acceptability of the offering did not depend upon- the parentage,, but 
upon the knowledge, of the officiating priest. 

The original distinction of colour which marked the Aryan race from* the indige- 
nous is- alluded to* once or twice* in the Vedic hymns,, because the event of the Aryan 
immigration* was still comparatively recent and the distinctions themselves may have 
still existed. But nothing of the land is ever alluded to in Manu's Code. Indeed, he 
cautions his Brahman friends against marrying" a girl with reddish hay" (III, 8) r 
because in a dark or Indian race, such as Manu and his fellow-Brahmans had become,, 
red hair is a disease, while to a white or blue-eyed, raee it is the sign of a healthy 

In the earlier days of Hinduism the great qualification required ina professional 
priest was, as we have already hinted, not the parentage of his mother, but an accurate 
knowledge of his father V art, or in other words intellectual ability — ability, that is> to 
grasp the endless- rules and remember the endless- texts necessary to the correct per* 
formance of the sacrifice. But intellectual quickness? as any one who has lived in 
India must know, is not a question of colour or race, and a Brahman of the olden time 
desiring to train a son in his own art would select, from among the sons born to him 
by his> various wives,, the one in whom he discerned the greatest aptitude, and no consi- 
deration of the tribe or class- to which the mother of such a son belonged would deter 
him from so making his choice. In fact, by the ancient custom of exogamous mar- 
riage a woman tooLthe rank of her husband as a matter of course, whatever her own 
origin might have been* Such was the rule in other parts of the world ; and if this 
had not been the case in India also,. Manu oould not have penned. such maxims as the 
following, all of whioh tell strongly, against the very principle of caste which it was 
the main objeotof his- code to- establish :-— " Whatever be the qualities of the man 
with whom a woman is united by lawful marriage, such qualities- she also assumes,, 
like a river united with the sea. Akshamiia, a woman of. the lowest birth, being, 
thus united with Vasishta, and Sarangi, being united with Mandop&la, were enti- 
tled to very high honour. These and other females of low birth have attained 
eminence in this world by the good qualities of their lords" (IX, 22-24). Vasishta,. 
as the reader need scarcely be reminded, was one of the most distinguished of the seven 
great Rishis or sages,, and one of. the great priests of the Vedic age, from whom 
Brahmans claim to have sprung. He, then, was the man who took a u woman of the 
lowest birth " as his wife, and by so doing raised her to very high honour. 

Enough has been said, then, to show that long before caste was established in 
India, the Aryan invader, to whatever class he might belong, was in the habit of 
taking the women of the country as wives, and that hence no caste, not even that of 
the Brahman, can claim. to have sprung from Aryan ancestors. But apart from all 
the written evidence thai can be quoted, in proof of this fact, the marked deteriora- 
tion of tone and sentiment, which is conspicuous in the later writings of the Vedic 


Digitized by 


( 78 ) 

age, is alone sufficient to show that the true Aryan race had by that time ceased to 
exist on Indian soil. 

In the earlier Vedio poetry the Aryans pourtray themselves in characters that 
might have fitted the Gaul, the Roman, the Homeric Greek, or the ancient Briton. 
They are a free and warlike race, fond of wine and the chase, boar-hunters, eaters of 
beef, and like the ancient Persians, with whom they were closely allied in blood, wor- 
shippers of the simple elements or forces of nature — the sun, the moon, the winds, 
the waters, the storm, Ac. But all this becomes changed two or three centuries later. 
The worship of the personified elements is hardened into a series of dry mechanical 
rites, which only a professional priest could perform. Vishnu, whom the Aryans 
bad honoured as the propitious and " all-permeating" Sun-god, is transformed into 
the smoky messenger of the sacrifice, and his elemental significance is forgotten. 
The repulsive and thoroughly non-Aryan Shiva, descending with matted locks and 
closed eyes from the mountains of the north, thrusts himself into the Brahmanic 
scheme, and becomes identified in name, but not in nature, with Budra, the Storm- 
god of the Yedic poets. The gloomy superstitions of the savage self-torturing ascetic, 
of whom this terrific divinity was the model, override the cheerful faith of the old 
Aryan warriors. The desire, so natural to the human mind, for a personal second 
life beyond the grave, to be enjoyed in the bright kingdom of Yama among the fore- 
fathers of mankind, is superseded by a morbid longing for personal annihilation by 
absorption into the World-spirit, Brahma. The widow, instead of being led away 
from the tomb of her deceased husbaDd, as she was in the old Yedic ritual, the text 
of whioh was wilfully corrupted by the new class of priests, is now made to burn 
herself alive on his funeral pyre. A five days 9 sacrifice of human victims — a rite 
altogether unknown to the Aryans who migrated into India— is performed once a 
year to the numerous gods of the changed pantheon. The Aitareya Brahmana> or 
book of liturgies, discloses to us a Brahman, by name Ajigarta, as the seller and 
butcher of his own son for the altar of sacrifice. The Mah&bhfirata alludes to Brahmans 
who not only sacrificed human victims but eat their flesh. In the same poem we find 
Bhima the hero thirsting to drink human blood, and Draupadi the heroine married to 
five brothers— a custom unknown to the Aryan invaders of India, but still practised 
by certain savage tribes in the south of India. Even the name of P&ndu, by which 
the heroes of the MahAbhArata were called, is suggestive of the savage tribe called by 
the very same name in the institutes of Manu. 

A similar deterioration of race is noticeable in the literature of the period. 
Speaking of the Brahmanas or liturgic treatises which succeeded to the early age of 
Yedic poetry, Professor Max Miiller (though himself an ardent believer in the Aryan 
ancestry of the Hindus) remarks:— "These works deserve to be studied as. the 
physician studies the twaddle of idiots and the ravings of madmen. They will disclose 
to a thoughtful eye the ruins of faded grandeur, the memories of noble aspirations" 
(Ancient Sanekrit Literature, page 889). The writer adds :— "Itis most important to the 
historian that he should know how soon the fresh and healthy growth of a nation can be 
blighted by priestcraft and superstition. It is most important that we should know that 
nations are liable to these epidemics in their youth as well as in their dotage." But possi- 
bly other writers might see the fact in a different light, not as the premature blighting 
of a fresh and youthful nation by superstition and priestcraft, but as the absorption of 
the Aryan blood into that of the pre-existing indigenous savage. Such is the view 
taken by Mr. Gough in his very interesting volume on the philosophy of the Upani- 
shads, a type of literature far superior to that of the Brahmanas, which have been so 
strongly condemned by Professor Max Miiller. At the close of his work Mr. Gough 
describes the Upanishads as "the loftiest utterances of Indian intelligence, but th? 
work of a rude age, a deteriorated race, and a barbarous and unprogressive community' *\ 
(page 268). Again in page 3, speaking of the material out of which the blood of the 
Brahman theosophist was qioulded, he observes :— " The greatest confusion has been 

Digitized by 


( T9 ) 

introduced into the popular study of Indian matters by the term Aryan. This word 
has been fertile in every variety of fallacy, theoretical and practical. Before the 
work of thought begins in India, the invading Aryan tribes have become Indo* 
Aryans or Hindus. They have been assimilated to and absorbed into the earlier and 
ruder population, whom they at first fought against as the dark-skinned Dasyus, and 
then made to till the soil and drudge for them as Sudras." 

But this is not all. A great prophet arose in about 500 B.O. who pre- 
tested against the whole theory of Brahmanical sacrifice and drew most of his con- 
verts from the lowest classes of the people. Buddhism, as we know, died out in 
India after contending for more than ten centuries with the rival creed, and the whole 
of India at last reverted to Brahmanism. But what could have become of the 
numerous orders of low-caste priests by whom this long-pending contest was sus- 
tained? It is not at all] improbable that they were won back into the earlier creed 
by the bait of admission into Brahmanbood, provided they brought their followers 
with them ; or, what is still more likely, that they themselves imperceptibly slided 
back into Brahmanism, retaining, however, the rank and title of priest and bequeath- 
ing the same to their children. Brahmanism has never been noted as a persecuting 
or missionary creed. But it has evinced, and still evinces, an extraordinary power 
of assimilating and absorbing every religious agency that crosses its path. We have 
seen already how, without any conscious effort, it assimilated hordes of aboriginal 
magicians or Ojhas, and thus created the large and powerful sect of Hindus called 
Baktas (see para. 122}. It is not less likely that it absorbed in the same way, one 
after another, the mixed brotherhoods of priests who represented the rival creed of 
Buddha. The great Brahmanical tribe known as Sarjupari or Sarwaria, and so 
called because they live to the east of the Sarju or Qogra, belong to the very 
districts in which Buddhism first arose and where it secured the largest fol- 
lowing. Mr. Garnegy is probably right in asserting that they were once Bud- 
dhists. They themselves have a legend which connects them with Ramohandra, 
the great king of Ajudhya, who is said to have transplanted them from the western 
to the eastern side of the Gogra. But the fact of such a legend having arisen 
implies that there was something unusual in their origin which the legend was 
intended to conceal. 

Local traditions in Oudh and the North- Western Provinces abound in tales of 
Brabmans being manufactured out of low-caste men by r&jas, when they (the r&jas) 
could not find a sufficiently large number of hereditary Brahmans to attend some 
sacrifice or feast For example, the Kunda Brahmaus of Partabgarh are said to 
have been manufactured by Rfija Manik Chand, because he was not able to collect 
the quorum of 125,000 Brahmans to whom he had vowed to make a feast : in this 
way an Ahir, a Eurmi, or a Bhat found himself dubbed Brahman and invested 
with the sacred thread, and their descendants are Brahmans to this day" (Oudh 
Gazetteer, I, 305). A similar tale is told of Tirgunait Brahmans in Vol. HI, 229 ; 
of the Pathaks of Amtara in Vol I., 365 ; of the Pandi Parwars in the Hardoi district ; 
of the large clan called Sawalikhs in the Gorakhpur and Basti districts, who have 
nevertheless assumed the high-sounding titles of Dube, Upadhay, Tiwari, Misra, Dik- 
shit, P6nde, Awaathi, and Pithak (North- Western Provinces Gazetteer, Vol VI, 351-2). 
Only about a century and a half ago a Luniya, or man of the salt-making caste, 
which ranks decidedly low, was made a Brahman by Raja Bhagwant Bae of Asothar, 
and this man is the ancestor of the Misra Brahmans of Aijhi (Gazetteer, Vol. VIII, 
Part III, 49). The writer remarks :— " Numerous Brahmans have been co*opted into 
the caste through the influence of the rajas of Argat and Asothar, when the latter were 
at the height of their power* To carry out this oeremony a number of Brahmans 
were collected, among whom the candidate was seated and with whom he fed. Hence- 
forth the man was known as a Brahman of the sub-division into which he had been 

Digitized by 


( 8* ) 

If such things could be done within recent times, what could have prevented 
• their being done during the last two thousand years, and especially during the long 
contest which prevailed between the rival creeds of Brahmanism and Buddhism, and 
who can now venture to assert that the Brahman of to-day is the lineal descendant of 
the ancient Aryan invader ? 

Mr. Growse is among those who oonsiJer that Brahman* witfi Kshatriyas are " in 
the main descendants of the early Aryan conquerors ;" and as regards the former he 
observes that " the strength ef a community which lays claim to any esoteric know-* 
ledge lies in its exclusiveness" (Mathura, page 414). But, so far as I can learn, the 
only class of Brahmans that lays claim to esoteric knowledge is that of the Ojha or 
sorcerer, and of all the BrahmanicaL sub-castes this is- the- one whose origin can be most 
clearly traced to aboriginal priests. 

Brahmanism in indigenous to India, From small beginnings it has gradually 
won over to its side almost the entire Indian race, and is even now continually gain- 
ing fresh victories. I believe thai one of the great secrets of its influence lies in the 
fact that its . professional expounders are one in blood, in character, and in sympa- 
thies with the general population. It is to me quite inconceivable, and opposed, I be* 
lieve, to all the teachings of history, that a race of over two hundred million soul* 
eould have been brought into the most abject spiritual subjection by a foreign priesthood. 

136» We oome finally to •the various celibate orders of devotees or religious 

mendioants, which, though they cannot be called castes in the 

proper sense of the term, represent one of the classes inta 

which the Hindu population is divided. One of them, too, has become a caste as well 

as an order,, and there ia another that threatens to follow. We cannot,, therefore,, omit 

them in this paper. 

137. For the origin of the celibate orders we must go back to the ancient dis- 
cipline of the Brahmans described in Manu's Oode. The reader is already aware that,, 
aocording to this Code,, the life of a Brahman,, or other " twice-born" man, was parcel- 
led out into four distinct stages, the two last of which were that of the Banaprastha or 
forest anchorite and that of the Sannyasi or ascetic. These two stages are the model 
on which the numerous orders of religious celibates have been founded. In. the time 
of Mann, it was only the twice-born oastes, and of these chiefly the Brahman, who. 
were entitled to enter into this holy state. But the modern religious orders are 
recruited from all castes, even the lower ones ;. and when the ceremony of induction 
has once been performed, all connection with the former caste is for ever cut off. The 
Brahman, once admitted into one of these orders^ oeases to be a Brahman, and ranks- 
no higher and no lower than any other member of the same fraternity. All members 
alike are supposed to have entered into the state of being dead to the world;, and in 
the spiritual life thus formed, no less than in the earthly grave, all men are equal. 
The first disciples of any new order that may be founded are actuated by the same 
intensity of faith that inspired the founder himself. But as time goea on, faith grows 
cold, and the enthusiasm of the first converts is not maintained by those who succeed 
to them. Thus has it been with most of the religious orders in India. The adults 
who now enter them are for the most part men who have become broken in fortune, 
or have no relations or descendants living, or have deserted their wives and families, 
or have lost their caste and cannot gain admission into any other. There is a Hindu 
proverb intended to signalize the decline of the religious spirit : : — 

" Ndri mux ghar sampatti nd$i\ 
Alund murdi bhae wnnydsi" 

" One whose wife ia dead, and who has lost his home and property, having shaved 
his head, becomes a Sannyasi" (enters a religious order). Another mode by which 
new adherents are gained is by picking up boys of a tender age, who have been dis- 
owned by their parents or have lost them, and training them to become disciples 

Digitized by 


( 81 ) 

(chela). This is the method which most orders prefer ; for it is easier to train children 
than elderly and broken-down men in the tenets and practices of the sect. 

138. For every order of devotee the ceremony of induction is substantially the 
same. The main difference lies in the fact that each order has its own gayatri or 
sacred formula recited at the time of initiation. The account given of the induction 
ceremony in Sherring's Hindu Tribes and Castes, Vol. I, page 256, is so very inaccu- 
rate as well as incomplete thnt I must here attempt to describe briefly what it is, so 
that the remarks which follow may not be misunderstood. In every order there is 
firstly the Bandprastb" or anchorite stage, which is considered incomplete and prepa- 
ratory, and seco" ' ^annydsi or ascetic stage, which is complete and final. For 
the indup*' A into each of these stages a separate ceremony exists. 

Th\ / two days and is as follows : — On the day fixed for the induc- 

tion (the \ a generally selected with reference to some periodical festival set 

apart in ho. A the deity to whom the order is specially attached), several promi- 
nent members of the order are invited to assemble at a certain place and witness the 
initiation. The man who has brought the disciple, and who is for this reason called 
his guru or spiritual guide, introduces him as an intending novitiate to the assembled 
members, and it is he who performs the ceremonies necessary to his induction. Holy 
water drawn from some sacred stream or tank, or (if such can be procured) water 
which has been already poured as an offering on the shrine of the tutelar deity of the 
order, is first thrown on the head of the novitiate in the presence of the congregation. 
He is then shaved by a barber, only a tuft, called shikha or chundi, being left on the 
top of the head. When the shaving is finished, he is congratulated and blessed by 
the members present ; sacred texts appropriate to the tenets of the order are whispered 
into the ear by the guru presiding ; and (unless he happens to belong to one of the 
twice-born castes, in which case he would possess a thread already), the janeo or Brah- 
manical thread is put over his left shoulder. At the same time a large sweetmeat is 
divided among the assembly, and after the guru has eaten most of his own share he causes 
the novitiate to eat the leavings — an act which renders it impossible for him to return 
to his former caste and binds him for ever to the guru as his disciple or chela. On 
this day he is considered to be a Brahman, and the main object of this part of the 
ceremony is to make him one. The title which he receives or holds in this capacity is 
that of Brahm&chari or religious student, thifr being the first and earliest of the four 
stages of a Brahman's life. On the second day the tuft is shaved off, not by a barber, 
but by the guru himself, and the sacred thread is cut to pieces and discarded. The 
object of this is to signify that the novitiate has left the status of Brahman and 
entered the higher one of Banaprastha or forest recluse. 

Thus far, however, the novitiate is only an anchorite and not yet a full-blown 
ascetic or Sannyasi. In this incomplete state he may remain, if he chooses, for the 
next thirty or forty years, though the intention is that the final ceremony which will 
make him a Sannyasi should follow soon afterwards. The preliminary ceremony is 
performed, as we have seen, by the guru ; but as the final one is performed by the 
anchorite himself, it rests with him to choose his own time for doing it. Faith, as 
we have said, grows cold after the first few generations of disciples are gone ; and 
hence most anchorites at the present day postpone the last aot till they are old enough 
to have outlived all their passions, because by that time it is easier to undergo the 
severer discipline and self-mortification required by the rules of the order. 

This final ceremony is called Vijaya Homa, which signifies " the oblation of 
victory." It consists of an offering of homa prepared after the manner described 
already in para. 112 and thrown upon consecrated fire. The throwing of this offering 
into the fire is intended to typify the entire dedication of the soul to the flames of 
ascetic devotion, the complete "viotory" over all earthly passions. Such is the 
efficacy ascribed to this rite that the bodies of those who have performed it are not 
burnt after death, as is the custom amongst all other classes of Hindus, but are either 


Digitized by 


( 82 ) 

buried in the earth or thrown as they are into some river. The Vijaya Homa is itself 
looked upon as a kind of cremation ceremony ; for the body of anyone who has under- 
gone it is supposed to have been reduced to ashes by the flames of his own devotion. 
This explains, too, the meaning of the word Goshayen, the name by which one of the 
greatest of the religious orders is called. It is derived from go and swdmi or shdyi, which 
writers have hitherto translated as " master of cows." But this gives no sense. Go is 
sometimes used in Sanskrit as a collective term for the five senses and the five corres- 
ponding organs. Hence the most probable meaning of Goshayen is " one who has 
mastered his senses." It is a synonym, in fact, of Sannyasi, but has become the proper 
name of an individual order. 

139. The number of religious orders in Upper India is very great Even in the 
ascetic state, when self-salvation is supposed to be the one pursuit of life, a Hindu is 
nothing if he is not a sectarian, attached to some particular deity or tenet in the mul- 
tifarious creed of Hinduism, and bent upon furthering the interests and influence of 
the sect to which he has given his allegiance. The census report is altogether silent 
on the subject of these religious orders and has not one word to say either about 
those of the Hindu or of the Muhammadan persuasion in section IX of the Prelimi* 
nary Dissertation on the " Religions of the People." The figures, too, recorded in 
form VIII are so obviously inaccurate, that no comparison between the results of the 
last and the preceding census could yield any useful result. The plan upon whioh 
I have arranged the religious orders of the Hindus is given below, but only the more 
important of them are named under their several headings : — 

I, — Followers of Shiva, the third god of the Triad — 





Sannyasi proper. 








II. — Followers of Vislmu, the second god of the Triad, in one or other of his 


Badha Vallabhi. 




Bhirite in name as well ai in character. 

h Yishnnite in character, bat not in name. 

III. — Followers of either Shiva or Vishnu, but only according to the teaching of 

some particular prophet, who showed the panth or right way to worship him :— 

Gorakhnithi -« 

Bharttari ... ... 

BaitaliBhat ... 

Harischaodi ... ... ... 

Bamavat , ## ... ... 

Kamanandi ... 

Charandasi ... ... ... 

Baid&spantbi .. ... ^ 

Kabirpanthi ♦.. „, 

Dadupamhi ... ... ... 

BadhaDpanthi... ... ... 

Udasi ... ... „, 

tf anakpanthi ... 

a Kan »•• ... ... 

8uthra ... ... ... 

Kukaphanthi ... ... ... ...J 

140. The deity to whom the mendicant orders originally looked as the type and 
founder of asceticism was the non- Aryan Shiva, and not the Aryan Yishnu. The cha- 
racter, under which the former first made his way into the Brahmanic pantheon as 
the third and last member of the Triad was that of the model ascetic, with closed 
eyes and bated breath, frequenting grave-yards, smeared over with the ashes of cow- 
dung, and wearing matted locks. Such was the character that he wore when he 
overthrew the altar of Daksha (see para. 123), and thus established his right to the 
title of Mah&dev, or the Great God. The greatest and probably the oldest of the 
ascetic orders, the Goshayen, is based upon the worship of this divinity. One of the 
numerous names by which he is called is Jogeshwar, or the Lord of Ascetics. If the 
followers of Vishnu have founded certain sects of devotees similar to those which bear 
the name of Shiva, this has been done more out of jealousy for the credit of their 

Digitized by 


( 83 ) 

own divinity, Vishnu, than because there was anything in the creed of Vishnu that 
prescribed the ascetic state. 

141. Every one of the above orders has one or more external badges (bhesh) 
by which it can be distinguished from any other. It is only thus that the public can 
know whether they are distributing their alms to the sect whose doctrines they prefer. 
Hence the Eftndu religious orders have been called by the generic name of Bh£sh- 
dh&ri or badge-wearer ; and this is a more appropriate designation than the name more 
usually given, " Sannyasi, " which for the reasons stated in para. 138 is applicable only 
to those anchorites who have entered the final stage. Another generic name in com- 
mon use is S&dhu, which is ambiguous, because it can be applied to any pious man 
who has not formally entered into the status of a devotee. 

The badges assumed by the various sects are much too numerous to be detailed 
in this place. Almost every sect has three different sets of badges: the tildk or 
device painted on the forehead and elsewhere "with sandal-wood powder, clay, Ac; 
the colour of the cloth worn round the loins ; and the staff or pot or shell or other 
kind of implement carried in the hand. Some too have a peculiar way of wearing the 
hair ; but in one respect all sects are alike, vie., that they never allow their hair to be 
cut. As regards the tilak every Shivite sect paints itself with the figure of a half moon 
thrice repeated, one under the other, or with some variation of this : this device is called 
tripund. On the other hand every Vishnuite sect paints itself with the symbol called 
ramanandi or some variation of it. The device consists of three parallel lines drawn 
perpendicularly, the two outside ones being joined at the base by * curved line which 
does not touch the middle one. The Shivite and Vishnuite sects are further distin- 
guished by the kind of rosary that each of them wears; for every religious mendicant, to 
whatever order he may belong, must wear a rosary of some kind to enable him 
to keep account of the number of times that he repeats die name of his deity or the 
words of the Gayatri or sacred verse peculiar to his order. The Shivite rosary is 
made of the seeds of a tree called mdraksh, or the " eye of Kudra " (Shiva), being so 
called because there is a mark on the seed shaped like an eye,which the Shivites 
regard as the symbol of the middle eye of their three-eyed god— the middle eye which 
is at present closed, but which he will one day open to destroy the world with fire. 
The Vishnuite rosary consists of beads made of the wood of the tulsi plant, this being 
the plant into which the nymph Tulsi was metamorphosed when she was pursued 
by Krishna, one of the incarnations of Vishnu. 

142. 8ome orders are further distinguished from others by the austerities which 
they practise or profess to practise. Some, for example, keep the right arm always erect 
in the air ( Urddb&hu). Others keep their faces always turned to the sky and never look 
dov?n[ AkashmuJchi). Others maintain a perpetual silence (Maunidasi). .Others touch 
nothing except at the end of a staff (Dandi and Tridandi). Others practise holding their 
breath, or standing on one leg, or keeping one leg fixed behind the neok, or standing 
between four fires under the scorching sun of April, May, and June, or standing all 
night in water in the open air in the months of December and January, or exposing 
themselves day and night to the rain of July, August, and September ( Yogi). There 
are some kinds of penance which are common to several different orders. The most 
severe kinds are those practised by the Shivites proper named in class I. The least 
severe are those of the Vishnuite Pantlds of class III. 

143. By far the most interesting of the above orders are those enumerated 
under heading III. The founders of these sects represent the various forms of modern 
dissent from the letter of Brahmanical' teaching. The relation in which they stand to 
the dominant priesthood might be compared with that of the prophets of Israel to the 
priests and Levites of the Mosaic law. They were reformers, and the chief aim of their 
teaching was to protest against the the claim of Brahmans to superior sanctity or to 
superior spiritual gifts. They came from various different castes, and some of them 
from very low ones. RaidAs, for example, was a Chamfcr or hide-skinner ; Kabir, a Eori 

Digitized by 


( U ) 

or weaver, and Sadhan, a Khatik or butcher. But Brahmanistn has been too strong 
for them, as it proved to be too strong for S&kyainuni himself some two thousand year* 
ago. The older a seco becomes, the more steadily does it relapse into the fetters from 
which its founder wished to emancipate the disciples. Speaking of the polity of the 
church of Rome, Macaulay describes it as " the very masterpiece of human wisdom # * * 
She thoroughly understands, what no other church haa understood, how to deal with 
enthusiasts * * *, The Catholic church neither submits to enthusiasm nor proscribes it, 
but simply uses it." The same might be said of Brahmanism in India; for the effect is 
not less conspicuous, though a less degree of conscious effort is employed in producing 
it. Even the Sikh sects, the five named last in tbe list, some of which are of very 
recent origin^ though they began by proclaiming a deism of the most liberal type, and 
disavowed any leaning towards either the Hindu Parameshwar or the Musulman Allah, 
are rapidly falling back into the old ways and taking colour more and more from 
Brahmanic teaching. A layman of any caste may profess allegiance to some parti- 
cular panth or u way " without at all losing, or desiring to lose, the caste in which he 
was born, and different men may follow different panths within the same caste. Thu& 
a Kayasfch may be a Nanak-panthi, or a Charandasi, or a Ramanandi, without ceasing 
to be a Eayasth. In Hindustan these panths have permeated the several classes of the 
community less widely than in the Panjab. But, so far as their influence has extend- 
ed, it may be said that, if Hindus are divided socially and industrially by caste, they 
are divided religiously by the panths or schools of modern reformers. 

144. Out of all the religious orders in Upper India there is at present only one, 
viz., the Goshayen, which from a celibate order, as it originally was and as it has still 
partially remained, has become a caste in the strict sense of the term. Writing of 
Gosh&yens Mr. Sherring says :— " Brahmans, Kshatriyas, Vaisyas, and Sudras may, 
if they choose, become Goshayens ; but if they do so and unite with the members of 
this fraternity in eating and drinking, holding full and free intercourse with them, 
they are cut off for ever from their own tribes. It is this circumstance which con- 
stitutes the Goshayens a distinct and legitimate caste, and not merely a religious order" 
{Hindu Tribes and Castes, Vol. I., page 256). Every word of this comment is beside 
the mark* All the religious orders, and not merely the Goshayen, are so constituted 
that any person from any caste who enters any such order is cut off for ever from 
his own tribe. In this respect all the religious orders are alike ; and the very cere- 
mony by which the novitiate is made to eat what his guru leaves at the time of 
induction entails forfeiture of caste and cuts off all possibility of return to it, as we 
have explained already in paragraph 13& What makes the Goshayen a caste, and not 
merely a religious order, is the fact that it has ceased to be celibate, while the other 
orders have remained so. In other words, it has openly admitted marriage among 
its rules or at least among its customs, and hence it can and does perpetuate and 
extend itself from within like any other caste, and not merely from without like the 
other religious orders. The Goshayen, therefore, is both a caste and an order : and in 
this respect, so far as I know, it has no parallel in India. It is a caste, because it 
extends itself by natural increase from within; and it is an order, because it admits 
new adherents from without and because many of its members are celibates. 

145. Goshayens have grown into a caste, because they had previously grown 
into a priesthood, and as priests had acquired property in land, houses, and temples, 
the possession of which modified the aims and character of the fraternity. The found- 
ers and first disciples of the order had no intention of serving as priests to the 
outside community, or in fact of doing anything else than to wander over the earth 
as celibates and lead the ascetic life of which Shiva, their patron deity, was the pat- 
tern. But the piety of the people compelled them to become priests in the temples 
of Shiva, whether they liked it or no ; for it is only in the temples of Shiva or his 
consort, Kali, that priestly functions have been assigned to them. As was explained 
above (see paragraph 123), it is only Brahmans of an inferior stamp who will accept 
offerings placed on the lingam or Shivite symbol ; and hence a class of men like 

Digitized by 


( 8* ) 

<Jo3hayens, who were believed to have special influence with Shiva, and therefore to 
be better qualified than any others to receive on his behalf the offerings made to him 
by the people generally, supplied a public want. Thus, if a Brahman of the PandA 
class conld not be found to take charge of a Shivite temple, some Goshayen was selected 
for the office ; and by the rule explained already in paragraph 1 23 a Goshayen so 
appointed became thenceforth the owner of the temple and acquired the right of 
bequeathing all the interests attaching to it to his own successors. By the same rule, 
too, a Goshayen might purchase such rights from a Brahman ; and in this way many 
more of the temples of Shiva fell into the hands of Goshayens and remained there. 
Other Goshayens again, not employed as temple -priests, but leading the model life of 
an ascetic, have been presented at different times with gifts of land and houses by the 
pious laymen of the community, partly with a view to presenting a costly offering to 
the Great God Mah&dev through the medium of his chosen servants, and partly for 
providing a permanent source of charity to the poor and needy, of which the Goshayens 
would serve as trustees and distributors. Having thus acquired property of a kind 
which could not be moved, and which therefore compelled them to live in certain fixed 
places, a large portion of the fraternity ceased to be mere wandering mendicants, 
begging their bread from place to place, as all originally did and as some are still 
doing. But settled habitations and the permanent acquisition of wealth lead naturally 
to the marital instinct and to the desire for heirs, to whom property can be bequeathed. 
Moreover, the custom of illicit unions with women of the lower classes had long been 
secretly practised. The postponement of the final ceremony of " Vijaya Horn a" led 
to a gradual relaxation of discipline, especially among the younger men of the order, . 
who had been initiated from early boyhood. Low-caste women were seduced from 
their houses, and others who had no homes received shelter and were retained as mis* 
tresses. When sons and daughters were born, the former might be initiated nnder 
the disguise of boy disciples. But daughters could only be given in marriage ; and 
as no outside caste could accept such girls as wives to their sons, it was necessary to 
find husbands for them within their own community. Thus marriage became at last 
an openly recognized rule or custom of the fraternity, and so from a celibate order 
they became an hereditary caste. 

146. The special function, then, of Goshayens, considering them as a caste of 
priests and not merely as a religious order, is that of serving as priests in the temples 
of Shiva, and less frequently in those of his consort, Kali. Mr. Sherring observes : 
" In this part of India they worship Vishnu, though in some other parts they seem to 
be devoted to Shiva" (Hindu Tribes and Castes, Vol. L, page 257). Such a remark 
implies a fundamental misconception of the facts. They worship Vishnu at times, 
not because they are Goshayens, but because they are Hindus. But it is the worship 
of Shiva which has made them what they are. It is only on the Shivar&tri, or the 
great annual festival of Shiva, which takes place in February, that candidates for 
initiation are inducted into the order. The text whispered into the ear of the novitiate 
on that occasion is the Rudri, that is, the Shivite GAyatri, and not the ordinary 
G&yatri pronounced in honor of Vishnu, the Vedic Sun-god, which Brahmans and other 
twice-born castes are expected to repeat daily at sunrise, noon, and sunset Again, all 
Goshayens trace their origin to Sankara Ach&rya, whom they worship as an incarna- 
tion of Shiva, just as R&ma, Krishna, and many others are held to be incarnations of 
Vishnu. Sankara Ach&rya, as is well known, was the great teacher, who came from 
Southern India (Malabar) in the eighth or ninth century A.D. and preached the doc- 
trine of Shiva in opposition to that of Buddhism, which in Upper India was then 
in an advanced stage of decay. The ten branohes or sub-divisions of Goshayens fancy 
that they are descended from the ten eminent disciples on whom his mantle descend- 
ed when he left the earth. Shivism is therefore the peculiar creed of the Goshayen 
order or caste, and it is only in the temples of Shiva that they can act as priests. Go- 
shayens of the Eanphata or ear-split class oan serve as priests and slayers of victims in 
the temples of Kali also. A conspicuous example of this is to be seen at the temple 


Digitized by 


( 8* ) 

of Devi Patan in the Gonda district, Oudh. la Assam, where the temples of Kali 
are especially numerous, 1 have heard that Goshayens are more prominent as temple- 
priests than Brahmans ; and even in Upper India they must he considered a rival, 
though still much inferior, caste. 

147. The marriage ceremonies of Goshayens are not the same as those of other 
Hindus There is a much less degree of formalism attending them, indicating, what is 
the fact, that the custom of marriage was foreign to the original aims and intention of 
the order. Probably by degrees the differences of ceremony will dieout, and Gosbayen 
marriages will be attended with as much fuss and expenditure as those of other Hindu 
castes. A change is coming over the whole spirit of the order. A considerable num- 
ber of Goshayens have, like Brahmans, become entirely secularized, all pretensions to 
be either priests or devotees having been abandoned. Like Brahmans, too, they have 
no speciality of function as a secular caste, but will take up almost any kind of work 
that comes to their hands, so long there is no ceremonial pollution attaching to it. 

148. There is one other celibate order which threatens ere long to become a 
caste, and by precisely the same process that has made the Goshayens one. This is 
the order of Bairagis, who hold about the same degree of influence and wealth among 
the Vishnuite orders that Goshayens hold among the Shivite. Thus far no such 
thing as marriage is openly recognized amongst them. But they have acquired vested 
interests in many of the temples And other places sacred to Vishnu or to the deified 
men and animals who are associated with his history. The great Hanuman Garhi at 
Ayodhya— the fort of the flying monkey-god who aided R&ma in his invasion of 
Lanka, which is visited every day of the year by pilgrims from all parts of India— is 
in the hands of Bair&gis. In fact the whole of Ayodhya, the city so sacred to the 
memory of R&ma and so endeared to the hearts of all JBLindus, is overrun by this 
grasping and mendacious order, who point to one house as the spot where Rama was 
born, to another as the house in which he played as a child, to another as the 
courtyard in which his father, Dasharatha, administered justice, and so forth ; and 
they exact a liberal fee for the information. The same kind of fate has attended 
many of the other cities or temples sacred to Yishnu or his incarnations, such as Bithor 
in the Cawnpore district, Brindaban in Muttra, Chitrakuta in B&nda, Misrikh in Sita- 
pur, Dwarka in Gujerat, and many more. Bair&gis have acquired, too, a proprietary 
fight in some of the temples of the sun ; for Vishnu in the Vedic age, before he had 
begun to make himself incarnate in beasts, fish, and men, was merely an impersona- 
tion of this luminary. At many of the Surajkunds or Sun-tanks in Upper India, to 
which patients go to be healed of skin diseases, as Naaman the Syrian went to the 
Jordan to be healed of his leprosy, an image of the Sun-god may be seen on the bank, 
with some Bair&gi seated besido it, ready to receive the offerings made by the visitors. 
In another way, too, Bair&gis are following in the footsteps of Goshayens* They have 
acquired large properties in land given them by pious laymen as offerings to Vishnu 
and for the benefit of the poor. The boy disciples whom they initiate into their order 
are often their own illegitimate sons, and it is to such disciples that they bequeath 
the lands given to them for a purpose so entirely different. Probably the day is not 
far distant when marriage will be openly recognized as one of the customs of the 
order, and the Bair&gis will then have become a caste like the Goshayens. 

149. The jealousy with which the different orders regard each other is best seen 
at the Kumbh-ka-Melas or twelfth-year fairs, when crowds of pilgrims assemble from 
all parts of Upper India to bathe in the Ganges at Allahabad or Hardw&r. It is deemed 
especially fortunate to enter the water when the sun is passing from the zodiac 
called kumbh or pot and entering that of mahar or crocodile ; and violent contests 
arise as to which order is to have precedence. "The rivalry of the Bairagis and 
Goshayens culminated in the last day of the fair in 1760 in a pitched battle, which 
terminated in the defeat of the former, of whom some 1,800 were slain. Again in 
1796 the Goshayens, venturing to resist the better-equipped Sikh pilgrims, were 

Digitized by 




— 67,079 


... SO 






♦.. 119,904 


•m 191,466 


( w ) 

defeated with the loss df five hundred men" (North- West Provinces Gazetteer, VoL II, 
page 291)- As the reader is aware, the Sikh and Bairagi are of the Yishnuite class, 
while the Goshayen is Shivite. At such battles the honor of their respective deities is 
at stake. The Goshayens and other Shivite orders claim the right of the first bath, 
because the Ganges is said to have sprang from the matted locks of 8hiva, when he 
first descended into the plains of India from the peak of Kail&sa* The Bairagi, Sikh, 
and other Yishnuite orders claim precedence, because bj their creed the Ganges 
took its rise from the sweat of Vishnu's toes. 

150.' The statistics to be gathered from the census report as to the religious 
orders are not a little bewildering. The names that come nearest to this description, 
together with the figures recorded against each of them, are as follows :— 

Goshayea — ... 

Bharti ••• »•« ••• 

Rangaswami ... ... 


Faqir ••• *»* ... 


The entries against the name Goshayen cause no difficulty : for as this has long been 
not merely a celibate order, but a marrying caste, no surprise need be felt at the 
existence of 51,180 females against 67,079 males. The remaining four names taken 
together include 124,387 males and 101,78.4 females. But if these four names are 
intended to stand for the orders of religious devotees, there should be no females at all, 
or only a very few, such orders being professedly celibate ; whereas the number of 
females recorded is almost equal to that of the males. If, however, these names are 
not intended to stand for the celibate orders, then these orders are not shown anywhere 
in the census report as they should have been* Perhaps faqir, being a Muhara- 
madan name, is intended to stand for the Muhammadan mendicant orders, amongst 
whom marriage is openly allowed. But if this is so, they ought not to have been 
included in a list which professes to relate exclusively to Hindus. Again, it is not 
understood why such obscure and little known orders as Bharti and Bangasw&mi, the 
one consisting of only 49 and the other of 126 souls, shonld have received separate 
mention, and no mention be made of orders which are ten times as important and 
numerous. Again, the description of occupations recorded against faqir, viz., " reli- 
gious ascetic, beggar, 9 ' appears to imply that common vagrants or beggars, professing 
no connection with religion, have been mixed up with devotees to whom begging is 
a religious duty. It is therefore impossible for me to do anything more than add 
up the figures recorded against Bharti, Rangasw&mi, Kaparia, and Faqir, and call 
them by the collective name of " other mendicant orders." This has been done in the 
accompanying tabular statement (group VII.) 

151. So far as I can lesfrn from persons likely to be well informed on the sub- 
ject, the total number of males belonging to the several Hindu orders, within the area 
of the North- West and Oudh, is not likely to be less than half a million. Individuals 
from these orders may be seen wandering from place to place in all parts of the 
country. Groups varying in number from 50 to 150 are collected together in 
monasteries called maths or akaras at places of pilgrimage, such as Benares, Allahabad, 
Chitrakut, Hardwdr, Muttra, Ayodhya, &o. Probably most of the Brahmans who 
are now established at these sacred places as river-priests or temple-priests, under the 
name of Gangaputra or Panda, are descended from the illegitimate sons of celibate 
devotees, who permanently resided in the local monasteries and received the alms and 
adoration of the people* 

152. General summary. — We have thus completed the brief survey of Indian 
castes, which it was the object of this paper to unfold. If we omit the " foreign 
races not returned by caste" and "unspecified," neither of which have anything 

Digitized by 


< 88 ) 

to do with the subject, the number of Hindu tribes and castes recorded in 
the census report iu alphabetical order amounts to 182. From the examination 
which has now been made this number has been reduoed to 101, no less than 
81 of those shown in the census report haying been found to be either synonyms 
for castes already recorded, or the same names differently spelt, or names of sub- 
divisions of certain castes, or merely the names of occupations belonging to no caste 
in particular. On the other hand, the names of twenty-one castes, of which no mention 
was made in the oensus, have been added to the original number. % Thus the net differ- 
ence between my own list and that given in the census report amounts to 61, 

153. Another result elicited by these enquiries is that the names which were given 
in the oensus report in alphabetical order have now been classified according to the 
function or occupation, by which each caste is distinguished, and on the basis of which 
it was originally formed. The -order in which the several -castes have been arranged 
in the appended statements is believed to coincide inversely with the order in which 
they came into existence — inversely because our method has been to commence with 
those which were the last formed, and therefore stand at the bottom of the social scale, 
and thence to make a gradual ascent to the caste first formed— the Brahman, which 
stands at the top. Odtside the series altogether stand the casteless tribes, the 
remains of the ancient aboriginal population, out of whose blood the whole series of 
Indian castes was gradually formed. Each caste or group of castes represents one or 
other of those progressive stages of culture which have marked the industrial develop- 
ment of mankind not only in India, but in every other country in the world, wherein 
some advance has been made from primeval savagery to the arts and industries of 
civilized life. The rank of any caste as high or low depends upon whether the industry 
represented by the caste belongs to an advanced or backward stage of culture ; and 
thus the natural history of human industries affords the chief due to the gradations 
as well as to the formation of Indian castes. Such in rough outline is the theory of 
caste advocated in these pages. A general summing up of the various castes as thus 
classified has been given already in paragraph 8, and a more detailed list may be seen 
in the twelve tabular statements appeuned to this paper. As the mode of classification 
here adopted and the theory on which it is based have not, so far as I know, been 
proposed by any previous writer, a few general remarks will now be offered in further 
explanation of the principles and causes through the operation of which the great caste 
system of India is believed to have been formed. 

154. In the first place, then, I think it has been dearly proved that the bond of 
sympathy or interest which first drew together the families or tribal fragments of 
which a caste is composed, and formed them into a new social unit, was not, as some 
writers have alleged, community of creed or community of kinship, but oommunity 
of function. Function, and function only as I think, was the foundation npon which 
the whole caste system of India was built up. This, too, is the view expressed by the 
Hindus themselves respecting the origin of their own institution. Ho doubt most 
of the accounts given in the sacred literature, respecting the origin of caste, are my- 
thical and absurd, if we take them in their literal sense and make no attempt to extract 
the kernel concealed in the deistic shell. But rationalistic explanations are not want- 
ing in Hindu writings, as the following examples taken from the Mahabarata show :— 
" tihrigu replied : — " There is no (natural) distinction of castes ; this world, having 
been at first created by Brahma purely Brahmanic, became afterwards separated into 
castes in consequence of men's works 9 " (Muir's Sanskrit TextB y Vol. I, page 144). 
Again in another chapter of the same poem, describing the condition of men in the 
Krita, or best age of the world, it is said of the four castes " that tbey had but one 
Yeda and practised one duty, but bad separate duties" (Vol. I, page 145) ; that is, all 
alike were obedient to the same law of duty, but, subject to that law, exercised separate 
functions. The author of the Manava Code gives in his first chapter the usual mythi- 
cal account of the origin of the four great orders or classes into which the population 
was said to be divided. He shows, as others had done before him, how the Brahmin 

Digitized by 


( 89 ) 

issued from the mouth, the Kshatriya from the arms, the Vaisya from the thighs, and 
the Sudra from the feet of the Supreme Being. But the mythe itself, if we strip 
it of its theistio colouring, affords the clearest possible testimony to the fact stated in 
the Mah&bb&rata, that " the world was separated into castes in consequence of works," 
and the author of the Code himself can be quoted as his own interpreter : — " For the 
sake of preserving this universe, that glorious Being (Brahma) ordained separate func- 
tions {pfithak karmanx) for those who sprang from his mouth, his arms, his thighs, and 
his feet" (I, 87). In the next four slokas he describes in some detail what these sepa- 
rate functions are, which I need not however repeat, as they are already well known to 
the reader;. He winds up his account with showing that the Brahman is sprung from 
the most excellent organ (the head and mouth), because " he is the first-born of castes 
and possesses the Veda, and is by rigt\t the chief of this whole creation" (I, 93). The 
stress laid in this and in many other passages that might be quoted, upon the fact 
that the Brahman was the "first born 9 ' of the castes, gives an unforeseen support to 
the theory advocated in this paper, that the Brahman was the first caste that came into 
existence, and that all the other castes were formed upon this model, extending gra- 
dually downwards from the king and warrior to the hunting and fishing tribes, 
whose status is not much better than that of savages. 

155. In the 10th chapter of the Code the author informs us that " several im- 
pure classes have been formed by intermixture of marriage " between the four so-called 
original castes — the Brahman, the Kshatriya, the Yaisya, and the Sudra. After des- 
cribing in detail the various crosses between the four original breeds, he says : " These, 
among the various mixed classes, have been described by their several fathers and 
mothers ; but whether their parentage is known or unknown, they can be discriminated 
by their respective occupations" (veditavyah suxt-karmabhih). Thus the fanciful 
attempt to deduce the inferior olasses of the population existing in his own day from 
intermarriages " in the inverse order" between the four original castes so called is 
supplemented by the more rational theory which distinguishes one class from another 
by occupation. 

156. The views thus expressed by the Hindus themselves as to the origin of their 
own institution are entirely borne out by the evidence collected in this paper. The 
reader will have observed that special pains have been taken, in the account given of 
each caste, to ascertain the etymology and signification of the name by which it is 
called : for the signification of caste names is the most trustworthy indication we could 
possess of the process through which the castes themselves have come into existence. 
An analysis of all the caste names given in the appended tables, which, omitting those 
of the casteless tribes and the religious orders which are not castes at all, amoant to 
exactly one hundred, brings out the following result : — 



Local or 




Hunting caste* 

Fishing „ 

Pastoral — ... ••• 


Landlord ... 

Artisans, lower ... •.. 

Artisans, upper 


Se/Ting ... ..• .•• 

Priestly — ••• 



































The reader will observe, then, that seventy-seven per cent, of the names of the 
various castes are functional— that is, neither tribal nor local nor religious. He will 
observe, too, that out of the seventeen tribal names all but five (and even these five 



Digitized by 


( 90 ) 

exceptions can be explained) belong to the hunting and fishing castes ; and he is already 
aware that these fraternities are tribes as much as castes. If, then, oar theory is cor- 
rect — that caste was founded on hereditary distinctions of function, independently of all 
questions relating to tribal origin or descent — the above result is exactly what we should 
expect to find. These hunting and fishing tribes, being the furthest removed from the 
arts and industries of civilized life, and therefore still hanging on the outskirts of Hindu 
society, do not possess any function at all (save that of merely hunting and fishing) 
as members of the Hindu industrial system ; and consequently there is scarcely any 
kind of work to which they will not*turn their hands, if they can get the chance. But 
as we ascend the social scale, we find that the functions of the several castes become 
more and more specialized as well as more refined j and thus in all such castes func- 
tional names have superseded tribal ones. 

157. If we look to the conditions of human society in its earliest stages, it is 
obvious that, in the hunting and fishing states proper, life is so simple and one-sided 
that scarcely any division of labour is required; and where there is no division of labour 
there can be no hereditary functions, such as exist in more advanced communities, and 
no distinct classes of men to perform them. As Plato remarked more than two thousand 
years ago, " a state only begins to exist when the individual ceases to be self-sufficing 
and needs the co-opeiation of many." In other words, it is only when settled, that is, 
agricultural communities have been formed, that certain hereditary functions, either 
secular or religious, can be acquired by certain classes of the population. When this 
point has been reached, the hunting, fishing, and nomad tribes, if they are not alto- 
gether absorbed, as most of them are certain to be, into the agricultural and other indus- 
trial classes, are expelled from the community as outcastes ; or if they desire to be 
retained as members of the community, but to adhere to their old occupations of hunt- 
ing, fishing, and cattle-grazing, they must be content to accept such conditions as are 
consistent with the welfare and convenience of the more advanced portion of the com- 
munity. This is exactly what has come to pass in India. The Ahirs, J6ts, Giijars, 
Oaddis, &c, all of whom were originally nomads,and whom I hav$ therefore classed as 
being "allied to the pastoral state," have now become largely absorbed into the agri- 
cultural function, though they have still retained many of the traditions and customs 
peculiar to the nomad state, and are still in the actual practice of nomadism, wherever 
agriculture has left them sufficient room. Again the Pasis, Arakhs, Bhars, Ac, all 
of whom not many generations ago were mere hunters and marauders, and whom 
I have therefore placed under the heading of " allied to the hunting state," have 
abandoned the migratory life of the original hunter and formed permanent homes on 
the outskirts of the villages and towns, where they devote themselves to peaceful and 
industrial pursuits, as far as their opportunities permit. But they still hunt and trap 
wild animals and birds in the neighbourhood of their huts, and are still waging a 
secret war against a society which has failed to assimilate them. These make up the 
last and lowest link in the Hindu industrial chain. Outside this chain, and claiming 
no affinity with it at present, are the Kanjars, Haburas, Banm&nush, &c, who still 
lead the wandering and predatory life of the unreclaimed savage. These are the out- 
castes of Indian society— tribes, or rather fragments of tribes, which have lagged 
behind the rest in the onward march of progress, while their brethren have advanced, 
at different paces, from one stage of culture to another, impelled and guided by the 
same laws as those which have governed the industrial development of mankind in 
every other part of the world. 

158. There are several minor castes at the present day which are still in process 
of formation, and the process admits of direct observation in all its stages. In the 
case of all these incipient castes it is some peculiarity of craft or industry, and not com- 
munity of religion or community of kinship, which supplies the motive force. One- 
good example of this has been given already in the case of the Sangtarash oi£ 
stonecutter (see para. 62). Regarding this and other similarly incomplete casters 

Digitized by 


( 91 ) 

Mr. Growse has written as follows : — " These partially developed castes are only 
recognized in some few districts and totally ignored in others. Thus Mathura is a 
great centre of the stonecutter's art ; but the men who practise it belong to 
different ranks, and have not adopted the distinctive trade-name of Sangtarash, which 
seems to be recognized in Aligarb, Hamirpur, and Kumaun. Again, in every market 
town there are a number of weighmen, who no doubt in each place have special guild 
regulations of their own ; but only in Benares do they appear as a distinct caste, 
with the name of Palledai$. So, too, at Sah&ranpur some fruit-sellers, whose trade, it 
may be presumed, has been encouraged by the large public gardens at that station, 
have separated themselves from the common herd of Kunjras and decorated their 
small community with the Persian title of Mewafarosh" {Mathura, 415-6). Many 
other instances might be named of castes which are in the course of formation, and 
which may or may not become eventually complete. There is the Peshiraj or stone- 
quarrier, on the sides of the Mirzapur hills, who seems inclined to separate himself 
from the parent stem of Ahir or cattle-grazer ; for it is in the neighbourhood of these 
stone-quarries that the Ahir finds, woods and pasture for his herds. There is the Raj 
or Maimer, who seems to be gradually developing into a caste of masons, distinct 
both from the Peshiraj and the Sangtarash. There is the half-formed caste of Beldar 
or navvy, which shows signs of detaching itself from the parent stem of Lunia or 
flalt-maker ; for as the Lunia is no longer allowed by Government to practise the 
ancestral art of extracting salt from the alkaline soils of the earth, the caste is now 
turning to other callings, such as well-digging, road-making, &c, and the beldar is 
one of the results of this change of function. There is the half-formed caste of Chap- 
parband or thatcher — an offshoot from the caste ot Kori or weaver, but an oftshoot 
which has not yet been able to separate itself from the roots of the parent tree. There 
*is the Khemah-doz or tent-sewer, who stands in a similar relation to the parent caste 
of Darzi or tailor. There is the partially-formed caste of Gadahla, whose special 
industry consists in driving and rearing the ass, but who belongs at present to the 
caste of Kumhar or potter. Thus in every example that can be quoted it is some 
peculiarity of trade or craft which supplies the root of the new plant, and the 
name assumed by the incipient caste is invariably descriptive of the industry from 
which the said caste takes its rise. The process of caste-formation which is thus 
going on before our very eyes is as good an indication as we could have of the pro- 
cess by which the castes already established were formed from the beginning. 

159. The same rule will be found to hold good if we examine the process by 
which individual men and families are seen to piss out of their own into some other 
caste. It is always through change of function that the first step towards change of 
caste is taken. A barber, for example, who has taken to carpentry (and the case 
which I have in mind is not an imaginary one), though he will not be able himself 
to change caste from N&pit to Barhai, will be able to do so in his descendants ; and he 
himself will go some way towards it by associating with Barhais, attending their 
annual festivals, observing the trade rules, learning the caste traditions, &c. If h e 
sticks closely to the trade, and if his sons and grandsons follow in his footsteps, they 
will certainly be called Barhais or carpenters by their neighbours ; for the term 
Barhai, like most other caste terms, stands for an occupation as much as it does for 
a caste. They, too, will give themselves out to be Barhais, whatever their actual caste 
at the time may be ; for if you ask a native what his caste is, be is very likely to tell 
you some word descriptive of his occupation— so closely are the two ideas intertwined 
in his mind. The grandson or great-grandson of this barber-carpenter, if his posi- 
tion is sufficiently prosperous (for everything depends upon this), and if he watches his 
opportunity and bides his time, will not find much difficulty in forming a matrimonial 
connection with some family of the carpenter caste ; and when this point has been 
reached, the change of caste from N&pit to Barhai has been completed. Migrations 
of this nature from one caste to another are perpetually being made. Thus while the 
castes remain, the personnel of castes is liable to constant changes, through change of 
individual functions. 

Digitized by 


< 9? ) 

160. We bold, then, that the grouping of men by castes or sub-castes is at 
Bottom the same thing as grouping them by crafts, occupations, or functions. It 
appears, too, that in the minds of the natives themselves it has never meant anything 
else. We do not concur in what has been lately represented by a writer of note, that 
the grouping of men by castes consists in " grouping them by their folk or their 
faith, by kinship or by worship," and that these two institutions " are the roots from 
which society has grown up all over India," " The essential characteristics/' as the 
same writer says in another place, " of a man's state of 4ife and position among his 
people, those which settle who he is and where he belongs, are his kinship and his 
religion ; the one or the other, sometimes both * * * *. If, now, having laid hold 
of these two facts, we look around us in central India and try to perceive how they 
have been worked out, we shall find the simplest and earliest expression of them in 
two institutions — the pure clan by descent and the religious order ; the brotherhood 
by blood and the spiritual brotherhood ; those to whom a common ancestry, and 
those to whom a common rite or doctrine, is everything-" It seems to us that the 
formation of castes cannot be ascribed to either of these sentiments, and that if we 
eliminate the element of a common function and substitute in its place that of a com- 
mon ancestry or a common worship or both, we strip the word " caste " of almost the 
only meaning that it possesses. 

161. We shall have something to say further on as to the relation in which caste, 
the more recent product, stands to tribe, the earlier one. But we may at once allude 
to a certain fact, which alone appears to be sufficient proof that castes did not, as many 
tribes have done, spring up out of the sentiment of a common ancestry. Had this 
been so, we should at least been able to know who the reputed ancestor was in each 
case ; and every such ancestor would have received some kind of periodical wor- 
ship from his reputed descendants, as the custom was and is amongst tribes. In- 
stead of this, however, we find that there is no caste which has an ancestor to worship, 
and that every caste pays regular homage to the tools, instruments) or other objects 
peculiar to the function out of which it sprang. Thus (omitting the tribes or castes 
" allied to the hunting state," which, as we have already explained,, have no specific 
function as members of the Hindu industrial community) we find that the boating and 
fishing castes (see group II B.) sacrifice a goat to every new boat before it is put 
into the water, and that at the time of the Diwali they make an annual offering, which 
consists of red powder, oil, a wreath of flowers, and sweetmeats, to every boat they 
possess. Similarly, all the pastoral castes (see group II C.) pay a kind of worship 
to their animals by rubbing ochre or red earth (geru) on their tails, horns, and fore- 
heads; this is done on the annual festivals of Diwali, Holi, and Nagpanchatni. 
Similarly all the agricultural castes who plough the fields (see group II D.), and those 
members of other castes who have taken to agriculture as their means of livelihood, 

pay worship to the plough on the day called A$dri } when the monsoon sets in and the 
work of cultivation is renewed. Sugar mixed with melted butter and a few grains of 
rice mixed in water are thrown upon the plough at the spot where it breaks up the 
first sod of earth, and thus both the earth and plough receive a share of the offering. 
The grain-heap is similarly worshipped in the months of March and October, before it 
is removed from the threshing-floor to the agriculturist's own dwelling. The Tamboli 
or betel-grower pays similar homage to the betel plant in October, before he begins 
to pick the leaf ; and in July, before planting the new crop, he does homage to 
the ground prepared for the purpose. On the great annual festival of Dasahra, which 
is especially sacred to Chattris, all men of this caste (see group II.E.) worship their 
weapons of war— the sword, shield, matchlock, and bow and arrows, and the animals 
used in war, the horse and the elephant. Artisan castes of the lower rank (see 
group III.A.) all worship the tools by which they practise their respective crafts, and 
chiefly on the Holi r the great annual festival of the inferior castes. The Bansphor 
worships the knife with which he splits the bamboo and cane ; the Chani&r or tanner 
worships the rdpi or currier's knife ; the Buukar or Kori the apparatus with which 
cloth is woven; the Teli his oil-press; the Kalwar an earthen jar filled with wine ; and 

Digitized by 






( 93 ) 

the Kumhar the potter's wheel. Artisan castes of the upper rank (see gronp III.B.) 
worship their various tools on the Diwali festival, which to the more respectable castes 
marks the opening of the new year; the Bangrez worships ajar filled with dye ; the 
Halwai or confectioner does honor to his oven by placing against it a lamp lighted 
with melted butter. The trading castes (see group IV.) invariably bring out their 
rupees on the Diwali and worship them as the instruments of their trade. Among the 
serving or' professional castes (see group V.), the Kahar or carrier pays worship 
to the bahangi (which has been anglicized into " bhangy") before placing it for 
the first time across his shoulder. The N&pit or barber does homage to his razor ; 
musicians to their instruments ; the Kayasth or writer caste to the pen and ink. 
The dates on which periodical worship is paid to tools, Ac, are the Holi, the Diwali, 
or both ; the former being New Year's Day to the lower, and the latter to the upper 
oastes. The last and highest of all the castes, the Brahman, observes the annual 
festival of Rakshabandan (which to the Brahman is New Tear's Day) by paying a 
Jioma offering with much solemnity to the new sacred thread (janeo or upavit), with 
which he then invests himself as the badge of his priestly dignity. We find, then, 
in India a completely organized system of fetish worship, and the fetish of every 
caste corresponds with the function to which the caste itself is attached, 

162. Not one of the castes of Upper India, so far as I can learn, claims des- 
cent from a common ancestor, though there are some few which have a patron saint 
or deity peculiar to themselves. Thus the Kayasth or writer caste has Chitragupta ; 
the Lohar or ironsmith has Visva Karma ; the Bhangi or scavenger has Lai Guru ; 
Kah4rs or fishing castes have Raja Kidar; and the agricultural caste of Kurmi 
pays especial honor to Vishnu in his incarnation of the tortoise (kurma). The wor- 
ship paid in each case to the patron saint or deity of the caste only proves more 
distinctly than before that function was the foundation upon which the caste was 
formed. For Chitragupta in the Hindu pantheon is the recording angel who writes 
down the good and evil deeds of man in preparation for the day of judgment, as a 
Kayasth in a court of justice takes down the depositions of witness. Viswa Karma 
is merely a Vedic name for the deity in his capacity of artificer or blacksmith. Lai 
Guru or Lai Beg (Hindu or Muhammadan) was simply a king of sweepers, around 
whose name some fables have gathered. Raja Kidar is merely a Hindi corruption 
of Khwaja Khizr, the Musalman god of waters, who showed Alexander the Great 
where to look for the well of immortality, and who, therefore, became the patron 
saint of castes whose occupation is with water. Kurma is the mythical tortoise which 
supports the earth, and is therefore specially deserving of worship by castes who live by 
the tillage of the soil. Thus in every instance that has been named, it is function, and 
function only, which, besides making the caste, has determined the character of its 
patron deity or saint. In all this there is no trace of the worship of a common ances- 
tor. The sentiment of kinship or a common ancestry is likely to spring up, it is true, 
among a body of men who have been previously united by a common industry and 
common connubial rights ; but this sentiment is the effect, and not the cause, of the 
formation of caste. 

163. The theory that a caste can be regarded as " a religious brotherhood," a 
body of men, " to whom a common rite or doctrine is everything," appears to be 
equally groundless. If the reader will again refer to para, 156, he will observe that, 
in the analysis of caste names there given, there are only two which carry a purely 
religious meaning. And if he will refer to the general head of " trading castes," 
group IV., h© w iH find that the two castes referred to are Vishnoi and Maheshwari. 
Neither of these is mentioned in the list of castes given in the census report ; and thus, 
if 1 had confined myself to that list and had not made a new one of my own, I could 
have affirmed that there was not a single caste among those under report which owed 
its origin to the bond of spiritual brotherhood. But these two exceptions, when we 
come to examine them, only serve to place the functional theory in a stronger light. 
The Maheswari caste, as its name implies, must have originally consisted of a body of 


Digitized by 


( 9* ) 

traders devoted to the worship of Shiva (Maheshwara). The only bond, however, 
which has held them together as a distinct social unit is the possession of a commoa 
function ; for many members have beoome Vishnuite in creed and many others Jain. 
If the caste had had no better bond of union than religion, it would have brokea 
up into three distinct parts long ago. The name " Visbnoi," as applied to the caste 
so called, is a misnomer ; for they are Musalmans almost as much as Yishnuites. 
"They worship," as Sir Henry Elliot has shown, " according to the Hindu ceremonial 
three times a day, and pray after the Musalman fashion five times a day. They keep 
twenty-eight holidays during the year and observe the feast of Bamai&n. They 
read both the Quran and Hindu Potbis" (Supplemental Glossary, Vol. 1, para. 43). 
Some Vishnois, too, have become Jains, like the Maheshwaris. Both of the above 
caste* are extremely unimportant ; but had their importance been greater, and their 
religious unity more complete than it is, two solitary exceptions out of such a large 
number of castes would not have been sufficient to invalidate the functional theory. 

164. In the analysis of caste names given in para. 156 I have abstained 
from including the names of religious orders on the ground that the said orders are 
celibate sects and not castes or marrying communities. But considerable stress 
has been laid, by the writer referred to in para. 160, on " the formation of sec- 
tarian castes under the complex working of the ideas of kinship and religion combi- 
ned ;" and Professor Mux Muller has been criticized for not having included " sec- 
tarian 9 ' among the headings under which Indian castes can be grouped. It has been 
alleged by this writer that " the various sects representing the several phases of Hindu 
schism, and founded by mystics or devout ascetics, who spiritualize the idolatry and 
rude superstitions of the vulgar, * * * have almost invariably ended by a new Brah- 
manic caste or sect ;" " that almost all the sects, except those allied to Buddhism and 
its satellite Jainism, have formed separate castes." The same writer has alleged 
that " in Northern India there are several of these purely sectarian castes whose 
origin can be historically traced back to a famous personage, often a good fighter as 
well as preacher, who is now the semi-divine head oentre of the caste. Within at 
least one of these castes the idea of affinity has woven during the last three or four 
centurioB a wonderful net-work of separate groups, deriving from the various clans, 
castes or families of the proselytes, who at sundry times and in divers places have 
joined the sect." But the Sikhs (for this apparently is the sect referred to) can in 
no sense be called a caste, nor has any caste arisen out of them. In all questions of 
inter-marriage the old restrictions and distinctions are observed as closely by Sikhs 
as by any other of the numerous sects of Hindus, and in weddings and on other 
domestic occasions the Hindu ritual is followed. In fact the expression " sectarian 
caste' 9 is almost a contradiction in terms. One essential characteristic of the sects 
founded by " mystics" is that ibey (the mystics) are celibate, and celibacy forbids 
the very principle on which the existence of a caste depends, viz., transmission fiom 
father to son. Out of all the thirty-four sects named in paragraph 139 (and the 
list there given was avowedly incomplete), there is only one, the Goshayen, which has 
ended in forming a caste, and it has accomplished this only through violating the 
vows taken by the members on entering the order. The Goshayens, so far as they 
are a caste at all (for many of the novitiates remain faithful to the vow of chastity to 
which they bind themselves), are a caste of bastards, and no ,one would make bas- 
tardy the basis of a theory for the explanation of Indian caste. It is worth noticing, 
however, that the Goshayen had no sooner acquired a function, viz., that of priest in 
the temples of Shiva or Kali, than he grew into a caste— that is, a marrying or non- 
celibate community. His case, then, so far from proving that castes can be of sec- 
tarian origin, only proves that they are of functional origin ; for it is Amotion, and 
function only, which has made him what he is. Even now, however, the Goshayen 
is not regarded as a caste by the general community, though he has partially become 
one in fact. Most natives still call him an atil — that is, who has " passed away" and 
become dead to the world. 

Digitized by 


( 95 ) 

165. Caste, then, is a purely secular institution, and religion has had nothing 
to do with it Within the same caste, and even within the same gotra or sub- division, 
many different phases of religious belief, popularly known as Panths or ways, 
may and do co-exist. Thus within the Kashyap gotra of the Eayasth or writer caste, 
it will be found that some families are Vishnuite, others Shivite, others Sfikta ; others 
again are followers of doctrines founded by the various modern reformers, such as Ka- 
bir, Charan Das, Nanak, Ac. ; yet all this diversity of belief causes no disruption of the 
caste as an endogamous unit. Jainism, the satellite of Buddhism, separated itself 
from the Brahmanic creed and ritual by a much wider interval than Sikhism or any 
other form of dissent has since done. Yet in the caste of Agarwal there are Jains as 
well as Hindus, and a Jain father ha* no scruple whatever against giving his daughter 
in marriage to a Hindu boy. The writer whose opinions on the formation of clans 
and castes we have been examining has given the following remarks respecting the 
alleged formation of castes on the religious basis : — " It appears that a religious body 
with some distinctive object of worship or singular rule of devotion, has usually 
(though not invariably) come to split off into a separate group, which, though based 
upon a common religion, constructs itself upon the plan of a tribe. The common 
faith or worship forms the outer circle, which has gradually shut off a sect not only 
from inter-marriage, but even from eating with outsiders ; while inside this circum- 
ference the regular circles of affinity have established themselves independently, just 
as families settle and expand within the pale of a half-grown kibe. Each body of 
proselytes from different tribes and castes has preserved its identity as a 
distinct stock, keeping up the fundamental prohibition against marriage within 
the particular group of common descent. But with some other group of the 
sect it is essential to marry ; and thus in course of time has been reproduced 
upon a basis of common belief or worship the original circle of a tribe, beyond 
which it is impossible to contract a legitimate marriage * * *. Several instances could 
be given of sects having gradually rounded themselves oft into complete castes, 
neither eating nor marrying with any beyond the pale." So far as I can learn, 
not a single example of this alleged process, by which a sect becomes rounded off 
into a caste with a distinct outer circle of marriage based upon a common faith or 
worship, has ever occurred in India, excepting only in the case of Goshayen ; and 
this sect, as we have shown, only grew into a caste through bastardy, that is, by 
violating the vefy principle on which the sect as a religious brotherhood was found- 
ed. Had caste been a religious and not an industrial institution, it would not have 
been possible for any caste to have survived, as many have done, the conversion 
of a large part of its members from Hinduism to Islam. A more radical change 
of feeling and conviction than what is implied in such a conversion as this could 
scaroely be conceived. Yet the hereditary transmission of functions has survived 
amongst those Muhammadans who are Hindus by blood and were formerly Hindus 
by religion, 

166. The hereditary transmission of functions from father to son, and the con- 
sequent distribution of society into various functional or industrial groups, have not 
by any means been confined to India. "Nothing indeed," says Comte, " could be 
more natural at the outset than that by domestic initiation, the easiest and most 
powerful means of education, employments should descend from fathers to sons ; 
and it was the only possible training in an age when oral transmission was the 
sole means of communicating conceptions" (Potitiv* Philosophy, Book VI, Chapter 8, 
page 198). In more than nine-tenths of the industries of India at the present day, 
oiml transmission, through the example and teaching given by the father, is the sole 
means, both among Hindus and Muhammadans, by which a boy can acquire his father's 
craft, and this is no doubt one reason of the hold which caste has had on the people of 
this country through all vicissitudes of creeds. Moreover, the possession of transmitted 
skill and the reputation acquired by the father as a oraftsmau are often the most 
valuable legacies which a father can bequeath to his son, and during the father's own 
lifetime the son is the cheapest and best helpmate that he can secure. Hence there 

Digitized by 


( 96 ) 

are few civilized or semi-civilized countries in the world in which caste, in the in- 
cipient stage at least, has not existed, and even in the backward races, wherever the 
germs of specialized industries can be traced, society at once begins to distribute 
itself in the hereditary groove. We will now quote a few examples in illustration of 
this position. 

The ancient inhabitants of Attica were divided into four hereditary classes, viz., 
priests, warriors, husbandmen (including herdsmen), and artisans. A similar divi- 
sion prevailed among the Ionians generally, of whom the people of Attica were 
one particular section. In fact class distinctions and class prejudices permeated the 
whole frame-work of ancient Greek society, and it was Clisthenes, the founder of 
the Athenian democracy, who abolished this system in Attica by substituting the 
qualification of wealth for that of birth and function. The Cretans, as Herodotus tells 
us, had a caste system of their own, something like the Egyptian, and he especially 
alludes to the agricultural and warrior classes. We hear of hereditary priests in 
Elis, Sparta, and Oela, and of hereditary musicians and painters in many other 
states. The population of Lacedaeraonia was divided into three great castes— the 
Spartiatoe or citizens of Sparta itself, who monopolized the functions of priest, ruler, and 
heavy-armed warrior ; the Perioiki, who constituted the tradesmen and artisans of the 
provincial towns and furnished the inferior infantry; and the Helots, who tilled the fields 
and did every kind of menial and manual work for the ruling caste. The Lacedaemo- 
nians were further sub-divided, as Herodotus tells us, into many minor castes, similar 
to what we find in India :— " Their heralds, musicians, and cooks inherit the trade of 
their fathers : the musician roust be the son of a musician, the herald of a herald, the 
cook of a oook. A man with a louder voice cannot come into the profession of herald 
and oust the herald's sons. Such are the customs of the Lacedasmanians" (Book 
VI., 60). Plato, in his ideal scheme of a republic, advocates the compulsory 
division of labour among six hereditary classes — Priests, Warriors, Husbandmen, 
Artisans, Shepherds, and Hunters ; and he quotes the example of the Egyptians 
as his precedent. It will be seen that this list of castes coincides exactly with 
the scheme on which I have myself classified the Indian castes in this paper, except 
that the Trading and Serving castes must be added to make the list of headings 
complete. As to what the Egyptian caste system precisely was, authorities differ ; 
but the general conclusion to be drawn is that the main divisions were very 
much what Plato described them to be, and that besides these there was a multitude 
of minor divisions, such as cowherd, swineherd, merchant, boatman. &c, as is the 
case in India. The Iberians (the ancient inhabitants of Spain) were, according to 
Strabo, divided into hereditary classes or castes consisting of kings, warriors 
priests, husbandmen, and servants or slaves. In Peru, as Mr. Prescott tells us, be- 
fore its indigenous civilization was destroyed, "there were certain individuals care- 
fully trained to those occupations which minister to the wants of the more opulent 
classes. These occupations, like every other calling and office in Peru, always des- 
cended from father to son. The division of castes was in this particular as precise 
as that which existed in Hindustan or Egypt" (Bistort/ of Peru, I, 143;. In Cuzco, 
the old capital of Peru, the priests of the Sun-god, the public registrars, the learned 
men, and the singers, all transmitted their functions to their sons. In Mexico as 
we are told by another historian, " the sons in general learned the trades of their 
fathers and embraced their professions : thus they perpetuated the arts in families, to 
the advantage of the State." Among our own ancestors there are traces of a ca*te 
system, or of something approximating to one, having once prevailed. The Anglo- 
Saxon population of England was divided into Ealdermen (the highest class of. 
nobles), Thanes or the inferior nobles, Ceorls or churls (yeomen;, and Serfs. 
Fortunately in the catholic church there was no hereditary class of priests, such as 
there is in India, to give fixity and permanence to this arrangement : but the 
minute rales in the old Anglo-Saxon laws defining the conditions under which a 
ceorl could become a thane, and a thane an ealderman, appear to point to a time 

pigitized by 


( 97 ) 

when the hereditary nature of function and status was more rigorously enforced. 
In the Fiji Islands carpenters form a separate caste ; and the skill displayed by these 
islanders in boat-building, above that of their neighbours, may probably be ascribed 
to the rule of transmitting acquired skill from father to son. In the same islands 
each tribe is divided into five classes distinguished from each other by marked di veri- 
ties of rank ; ©{#., kings, chiefs, warriors, landholders, and dependants or slaves (Fiji, 
by J. H. DeRicci, page 67). In Abyssinia the people are divided at the present 
day into four great hereditary classes— the military and nobles, answering to the 
Indian Chattri or Rajput ; the sacordotal class, answering to the Indian Brahman ; the 
cultivators of the soil, answering to the Indian Kurmi, Ac. ; and the traders, answering 
to the Indian Baniya. Besides these there are a few tanners, saddlers, and black- 
smiths (Land of Prester John, by J. C. Hatten, p. 144(F). Something very like caste 
exists at the present day in Madagascar. The population is divided into the four great 
classes of nobles, Hovas, Zarahovas, and Andrvas or slaves ; no intermarriage ever 
takes place between them. As among the Chattris of India, so among the nobles or 
royal clans of Madagascar " their aristocratic lineage is no bar to their earning their 
bread as artisans and tradesmen." In the same island we are told that the middle 
class consists of the Hova freemen, divided into numerous tribes and families, each 
intermarrying only within its own circle (Madagascar, Past and Present, by Miss E. M. 
Clarke). In a recent article called " Russia Revisited," by Canon McColl, this writer 
points out that one of the first things needed for the amelioration of the people is 
" the reform of the caste system which divides society into classes separated from 
each other by chasms very difficult and often impossible to pass, .Nobles, merchants, 
clergy, are separated by rigorous rules which practically confine each class to its own* 
territory. The nobles again are subdivided into the great nobility and the little 
nobility, hereditary nobility, and personal nobility. The clergy are divided into black 
and white— that is, those who must not, and those who must, be married. The mer- 
chants are classified in three categories. Then there is the innumerable host of Chino- 
vinks— that is, all the functionaries of the civil service. The army may be regarded' 
as another caste. This caste system breaks up the unity of the nation." With a few 
changes of words this description would give Br fairly correct view of the social con- 
dition of India at the present time. 

167. Enough, then, has been said to shew that not only in India, but in almost 
every other country similarly advanced, the several arts and industries, springing up 
out of those natural divisions of labour which are necessary to the convenience of 
settled communities, were at the outset and for several centuries following hereditary, 
and that the first great advance made by nations, after they had emerged from the 
hunting and nomad states and begun to adopt the modes of life coeval with the agri- 
cultural state, was made in this groove. But so long as the hereditary transmission 
of functions is simply natural and spontaneous, and no artificial rules are invented for 
its support and continuance, no such thing as caste, in the proper sense of the term, 
can be said to exist. Mere heredity of function is not the same thing as caste ; or if 
it can be called by such a name at all, it is caste only in the incipient stage — the foun- 
dation without the superstructure. Before anything like a complete conception of 
caste can be said to have been formed, there are two great questions to be asked and 

(1) In what respect does an hereditary industrial class fall short of a caste 
properly so called. ?. 

(2) In what relation does caste stand to the earlier organization of the tribe- 
which it superseded ? 

168. With a view to obtaining an answer to the first of these questions, we must 
go back once more to the time when the Brahman, "the first born of castes" — the 
model upon which all the other castes were subsequently formed— did not exist as a> 
easte, but mearly as a professional priesthood, 


Digitized by 


( 98 ) 

The mode in which a chief or any other personage of Vedic times sought to 
propitiate the gods was by inviting them to come down and lick the blood of animals 
slaughtered npon the altar, to smell the odours of flesh roasted on consecrated fire, 
to imbibe the fumes of the intoxicating Soma, and to hear praises sung in their honor 
by the bards and sages of the day. So long as the ceremony was simple, no profes- 
sional priest was required. Even when it had become more lengthy and complicated, 
a man of any rank or station in life was authorized to perform it, provided he was 
master of the details necessary to its efficacy. The great example of Visvamitra, 
who was descended from a long line of kings both of the Bharata and the Kasika 
families, and yet was appointed domestic priest (purohita) to King Sadas and sacrific- 
ing priest (hotri) to King Harischandra, shows that men of royal parentage conld not 
only perform sacrifices for themselves without the help of professional priests, but 
could officiate as professional priests for others. 

But a complicated ceremonial, the efficacy of which depended upon the rigidly 
exact performance of all its minutest details, could not but lead to the creation ; of a 
highly-trained priestly class, and such a class would necessarily become hereditary, for 
there was no other way in which an art so difficult as that of performing a Vedic sacri- 
fice could be acquired except by a long course of instruction under a father's guidance. 
Moreover, initiation under a competent spiritual guide was the only guarantee to the 
public, that the man offering his services as priest possessed the authentic tradition 
of those magical texts, gestures, and intonations, whose virtues called down blessings 
on the head of the worshipper and made his oblations acceptable to the gods* Among 
the nine qualifications laid down in one of the Sutras, or books of aphorisms, as 
necessary in a professional priest, the first and most important was that he must be an 
araheya — that is, able to trace an unbroken descent for ten generations in the family 
of some Bishi or inspired sage (Muir's Sanskrit TexU, Vol. 1, 513). 

But the hereditary transmission of a function, whatever the nature of the func- 
tion may be, depends upon the father only ; for it is the father, and not the mother, who 
acts the part of instructing the son in the hereditary craft. No question need arise 
as to the race, stock, or class to which the mother belonged. And in point of fact, no 
such question was raised in the early days : for by the ancient law of marriage (as the 
M&nava code itself testifies), the woman could rise to the rank of the man, whatever 
her own rank might have been, if she was lawfully married to him (Manu's Code, IX, 
23). Of all the Rishis or inspired sages of the Vedic age, Vasishta is the one in whose 
history the future position of the Brahmans is most distinctly foreshadowed. But 
Vasishta is described in the Mahabharata as the father of a hundred sons, and as 
having begotten a son to King Kalinashapads, whose queen had had no issue through 
her legitimate husband* He is also described in the Manava Code as having taken to 
wife Akshamala, " a woman of the lowest birth. 9 ' Yet there is no Rishi to whom 
a Brahman of the present day would more gladly trace his origin than to Vasishta. If 
we take the rule of " ten generations 9 ' quoted above, and if we suppose an extreme case, 
as we are authorized in doing, that in each of these generations the father was married 
to some woman of the non-priestly class, then the amount of Rishi blood inherited by 
the tenth in descent would be in the proportion of only 1 to 512, which is as good as 
saying that it would be reduced to nothing. And yet all the traditions, rules, and 
divine precepts of the inspired sage might have been handed down to his tenth descend- 
ant quite as faithfully, as if the nine mothers, through whom the series was continued, 
had been of the purest Rishi blood. In the Vedic age so little was a priest or sage 
prevented from marrying a woman outside his own class, that even Indra, the highest 
god of the Vedic pantheon, is said to have allied himself to a Danavi — that is, a woman 
of the aboriginal race. The legendary discussion between Yudisthira and the serpent, 
recorded in the Mahbharata, respecting the origin of Brahmans, draws a striking 
picture of the freedom that once prevailed in the choice of wives : — " The serpent 
remarked—' If a man is regarded by you as being a Brahman only in consequence of 
his function, then birth is vain until action is shown. 9 Yudisthira replied— 4 most 

Digitized by 

Google [ 

( 99 ) 

sapient serpent, birth is difficult to be discriminated in the present condition of man- 
kind on acconnt of the confusion of all classes. All sorts of men are continually be- 
getting children on all sorts of women. The speech, the mode of propagation, the 
birth, the death of all mankind are alike' " (ttuir's Sanskrit Texts, I, 137). 

169. Mere hereditary of function, then, is not sufficient by itself to constitute 
a caste. This can only be accomplished when restraint is put upon the freedom of 
marriage by establishing and enforcing the rule that no one can inherit the name and 
function of his father, unless his mother as well as his father was born with the same 
class. The two things, then, which constitute caste are heredity of function plus 
limitation of marriage. The principle of choosing a wife from some family enjoying 
an equal rank and exercising the same function is so natural, that it must have been 
growing into a custom long before it was made compulsory by social penalties. But 
so long as the opposite practice wa* permitted, the dignity and privileges attaching 
to Brahmanhood were at stake : for men could not continue for ever to accept such 
sophistry as the following :— " Sages who had begotten sons in an indiscriminate 
way conferred upon them the rank of sages by their own austere fervour " (Sluirs 
Sanskrit Texts , Vol. I, 132). If, then, the dignity of Brahmanhood was to be 
secured against all risks, it was necessary to lay down and enforce the rule that no 
one could be considered a Brahman, unless he was born one on both sides of his 
parents. Such was the rule insisted on by the author of the M&nava Code. " In all 
classes," says the Code, (X., 5) " they, and they only, who are born in a direct order of 
wives of the same class, and virgins at the time of marriage, are to be considered as 
the same in class with their fathers." One of the chief characteristics of Indian caste 
is contained in the words italicized. If, then, I were asked to define caste in such a way 
as to discriminate it from a mere hereditary olass, I should describe it as a trade-onion 
or professional group, of which no one is entitled to exercise the function and enjoy 
the privileges, unless he was born within its ranks from both parents* The word 
" caste" is derived from the Portuguese casta, which signifies "breed." As purity of 
birth from both parents is the differentia of caste as distinct from class, a more 
appropriate word could not have been introduced. 

170. The effect of such a rule as applied to Brahraans fand it was only for their 
benefit that the rule was made, Brahmans themselves having of course invented it), 
was that the essence of Brahmanhood consisted henceforth, not as it originally did in 
a personal qualification, that is, in the possession of brahma or sacred knowledge, but 
in purity of descent from both parents. Henceforth every man so born was entitled, 
by the mere claims of birth, to all the privileges and honors due to a professional 
priest, whether he did or did not discharge the priestly office or possess the sacred 
knowledge for which alone such privileges were originally conferred. Henceforth no 
one except a person so born, even if other men born of a different parentage might 
be more competent, could be allowed to exercise the priestly office for the public or 
receive gifts in return, " Whether ill or well versed in the Veda," says the Mah&- 
bhdrata, " whether untrained or trained (in sacred science), Brahmans must never be 
despised, like fires covered with ashes. Just as the fire does not lose its purity by 
blazing even in a cemetery, so too, whether learned or unlearned, the Brahman is a 
great deity. Cities are not rendered magnificent by ramparts, gates, or palaces of 
various kinds, if they are destitute of these excellent Brahmans " ( Vanaparvam, 
13,436). It was nothing but the most sordid and selfish ambition that led the 
Brahmans thus to constitute themselves into a caste, and in the rules laid down in 
the M&nava Code for defining the relations in which this caste was to stand to the 
other classes of the community, we are struck with the utter absence of reci- 
procity—the only element which could have imparted some degree of morality to 
their pretensions. Provided he married a Brahmani first, a Brahman might take con- 
cubines or inferior wives from the Kshatriya, Vaisya, and Sudra classes; but only 
the children of the Brahman wife could inherit the father's status, and no Kshatriya, 
Vaisya, or Sudra might marry a Brahman's daughter. Again, if the Brahman were 

Digitized by 


( 100 ) 

not able to Kve by bis own profession (that is, by sacrificing for others, teaching sacred 
science, and receiving gifts in return), he might take up any of the callings belonging 
to the other classes, provided they were such as not to interfere with his dignity. As 
the calling of a Kshatriya or warrior was next highest in rank to his own, he was 
directed to choose this in preference to any other, if his own calling did not maintain 
him (Manu X., 81 . But though he might bear arms and kill a fellow-countryman 
on the field of battle, he was forbidden to do anything so mean as to plough the 
earth with his own hands, and an hypocritical scruple was set up on the ground 
u that the iron-mouthed pieces of wood not only wound the earth, but the creatures 
dwelling in it " <Manu, X, 84). Again, supposing that he was rich and could dis- 
pense with the earnings of his own or any other calling, he was instructed to refrain 
from acting as priest or teacher to men of other classes who might require his ser- 
vices, because it was more dignified for a man of his parentage to stand aloof from in- 
ferior mortals. Hence, as no one except a Brahman could render such servicer, 
every other class in the community might go to perdition for all that the Brahman 
cared. But if the Brahman condescended to act as priest for the other classes, he 
must be liberally rewarded in return ; " for the organs of sense and action, reputation 
in this life, happiness in the next, life itself, children, and cattle, are all destroyed by 
a sacrifice offered with trifling gifts to the priests" (XI, 40). On failure of heirs the 
property of other classes escheated to the king, but that of Brahmans is reserved 
for men of their own caste (IX, 18«-9). A king, " even though dying of hunger," 
must never take taxes from a Brahman, but a Brahman, if he is in want, must always 
be fed by the king (VII, 133). If a Brahman was guilty of a capital offence, he was 
not to be put to death as other criminals were, but simply told 1 to depart elsewhere 
* with all his property secure and his body unhurt" (VIII, 380), A Brahman was 
never to be smitten " even with a blade of grass." " A twice-born man who touches 
a Brahman with intention to hurt him shall be whirled about in the hell of Tamisra 
for a hundred years"' (Manu, IV, 165), Offerings intended for the gods had 
much better be given to Brahmans, as the latter were visible deities and were 
certain to accept them, whereas there was an element of doubt as to whether 
such offerings reached the unseen powers : "an oblation in the mouth of a Brahman is 
far better than offerings to holy fire : it never drops, it never dries, it is never con- 
sumed" (Manu, VII, 84). Hence, as I have shown in another place, one of 
the chief functions of a Brahman at the present day is to eat at another man's 
expense. But no Brahman is ever expected to give a banquet to any but men of his 
own class or caste. 

171. "When the IJrahman had thus set the example of forming himself into an 
exclusive andhighly privileged caste, the other classes in the community were compel- 
led to take what precautions they could for securing such privileges as were within 
their reach ;. and they did this not merely in self-defence, but in imitation of a class o£ 
men whom they had been accustomed for centuries to regard with the deepest vene- 
ration. If Brahmans had been a celibate order, like the Roman Catholic priesthood 
in Western Europe, or if they had been held in as little respect as are the parochial 
clergy of Russia, amongst whom marriage is compulsory as it is amongst Brahmans,. 
the example which Brahmans gave of setting up caste barriers,.so as to secure their own 
monopoly against all outsiders, might have had no effect upon the general structure 
of society. The caste system which has now penetrated into every corner of India 
might in that case have never come into existence, or would have remained in the 
incipient and harmless stage, by which a son naturally inherits the craft or function 
of his father : and India would then have been spared the degradation and disunion, 
which the Brahman has inflicted on her. But the influence of the Brahmans, at the 
time when the M&nava Code was compiled, was overwhelming ; and this, added 
to the naturally superstitious character of her. own people, has wrought her ruin. 
u From priority of birth, from superiority of origin, from a more exact knowledge o£ 
Sacred science, and from a distinction in the sacrificial thread, the Brahman is the 

Digitized by 


t 101 ) 

lord of afll the classes" (Mann, X, 3). As in the Hindu family " the eldest son isin this 
world the most respected," and all other members of the honsebold are taught " to 
behave to him according to law, as children should behave to their father ;" so in 
the family of castes the Brahman, claiming priority of birth, superiority of origin, and 
a more perfect knowledge of sacred science, was the " first-born " son, to whom all the 
other classes looked for guidance, the oracle from which they drew their own inspiration. 
The principle of caste arrogance, once set in motion by the most influential class in the 
community, has been extending gradually downwards from the Brahman to those im- 
mediately next him in rank, till it has at last taken possession even of those inferior 
and backward classes who had no privileges or functions that are worth defending. 
A multitude of castes as thick as a cloud of locusts were thus called into existenoe at 
the bidding of the Brahman's wand? — 

As when the potent rod 
Of Am ram's son, in Egypt's evil day, 
Waved round the coast, up called a pitchy cloud 
Of locusts, warping on the eastern wind. 
That o'er the realm of impious Pharoah hung 
Xiike night, and darkened all the land of Nile. — Paradise fort, Book I. 

172. One of the best proofs we could have that the Brahman was the founder of 
the caste system of India, or at least the model upon which all the other castes have at- 
tempted to form themselves, is the constant recurrence of the number seven. Brah- 
mans of all kinds, aecording to their own assertion, are descended from the seven 
great Rishis or sages who are said to have laid the foundations of sacred science in 
India. The very lowest castes will tell you that their own communities, too, are made 
up of seven sub-divisions. Even Kanjars, who are still outside the pale of caste, will 
give you the same account of the organization of their savage community. Yet there 
is scarcely a caste or casteless tribe in India which is agreed as to what its seven 
sub-divisions are .; for the same caste or tribe will give you one list in one place and 
another in another. The truth is that this cotnpact scheme of seven, which almost 
every caste or tribe is so fond of repeating, has no foundation in fact. But these 
various tribes and castes could not have conspired togther to assert a common fiction, 
if this fiction, almost the only point on which they do not quarrel, had not been drawn 
from a common source— the Brahman. 

173. Two things, as we have shown, were necessary to the formation of a caste, 
viz., (a) some speciality of function, which served as the bond of union among the 
members and formed them into a distinct industrial unit ; and {b) the limitation of the 
right of membership to those who were born within its ranks from both parents, 
which made them a distinct social unit. A large number of industrial classes, possess- 
ing some hereditary function, was in existence long before the Brahman caste was 
formed ; for the Brahman himself was merely An hereditary priest at the outset 
The M&navaCode, though it maps out the population of the country on the plan of the 
old semi-mythical division into Brahman, Kshatriya, Vaisya, and Sudra, adds that 
there was a large number of other industrial groups, sprung " from the intermixtures of 
these classes by their marriages with women who ought not to be married ;" and these 
groups are alluded to incidentally in various parts of the code as potters, fishermen, 
physicians, usurers, cattle-grazers, boatmen, bards, oilmen, spirit-distillers, bowmak- 
ers, shepherds, barbers, goldsmiths, silversmiths, actors, basket-makers, washermen, 
sweepers, weavers, 4eather-outters, musicians, carriers or porters, carpenters, &c. In 
fact it would be quite possible to classify the industries alluded to in the M&nava Code, 
including of course those of the Kshatriya and Brahman, on the same plan as that on 
which the castes of the present day have been classified in this paper ; and in some cases 
even the names would coincide. All that was needed, then, for the conversion of each 
of these classes into castes was that it (the class) should follow the example of Brah- 

mans, " the first born of castes 1 ' and that of Eshatriyas, the second born, by establish- 
ing the same rule within its own ranks which the Brahman and Kshatriya had 


Digitized by 


{ 102 } 

already done in theirs ws. y that all intermarriage with outsiders should cease, and that 
mo one- should be entitled to the privilege of membership, unless his parentage was- 
of the approved type; There is therefore no- mystery about the origin of Indian* 
eaates, if we will only acoept two facts, (1} that oaste wae based at the outset oik 
some unity of function* and not, as some writers have affirmed, on unity of creed" 
or unity of kinship,, the several functions' themselves having been based on those- 
matural divisions of labour which* following each other in an* ascending scale from 
one stage of oulture to another, have marked the industrial development of mankind 
in every part of the world ; (2) that the Brahman had acquired an overwhelming as- 
cendancy over the- mind* of all the other clksses of the population^ and that these 
classes, being intensely ignorant and superstitious, had no alternative left to them 
but to follow the example of their priests.. The explanation of Indian caste, then,, 
centres in two great facts ; one general and scientific, vie., the progressive develop- 
ment of the various arts and industries acoording to certain fiked laws, which have 
been substantially the same everywhere t and the other, special and historical, viz ,. 
the extraordinary influence acquired by Brahmans over the ignorant and superstitious* 
masses. It is futile to attempt, as some writers have done,, to explain the origin of a* 
great social phenomenon, such as that of Indian* caste, on* purely general grounds,, 
and without faking into account those idiosyncracies of custom* and sentiment which* 
moulded the Indian character and marked" the career of the nation in the preceding: 
centuries : for man is- pre-eminently an historical beings shaped and fashioned into what 
he is by the accumulated influences handed down to him by every generation that' 
has gone before* Every nation has* had its* industrial development, and the process 
of development has been substantially the same in all; But what nation in the world* 
except India haa had its Brahmans? " Through the power of the Brahmans," says the 
llah&bhfirata, " the demons were thrown prostrate in the waters (of the abyss) ; by 
the favour of the Brahmans- the gods inhabit heaven. The void cannot become matter ;: 
the mountain Hi ma vat cannot be shaken ; the Ganges cannot bedammed ^tbe Brahmans 
oannot be conquered by any one on earth. The world cannot be ruled in opposition to* 
the Brahmans ;. for the mighty Brahmans* are the deities even of the gods. If thou- 
desirest to rule the sea-girt earth, honour them* continually with gifts- and with service"' 
(fAnuddsanap) 2H>0ff). Priestcraft has- done more to enslave mankind than the swordi 
of the most merciless oppressor. For a tyrant may be expelled in a day - r but the mindi 
once darkened by superstition lose* the power of vision and forgets its own blindness. 
Brahmans- have reigned in India for more than twenty centuries^, and no one cam 
foresee the day when their kingdom, will come to an end. Even practical Home was* 
at one time in. danger of falling; under the sway of a- dominant priesthood, —the rock* 
on which the foss practical India struck. There, as in India,. " the neglect or faulty 
performance of the worship of each god revenged itself in the corresponding occur- 
rence ; and as it was- a laborious and difficult task to* gain even a knowledge of one's* 
religious obligations, the priests* who were skilled in the law of divine things and 
pointed out its requirements— -the pontifibe*— could not fail to attain an extraordinary 
influence" (HommwnT* History of Rome, Vol 1). Prom that rock Borne was saved: 
by the nonfhereditary character of the priestly office, the indomitable enterprise of> 
her people,, their intense patriotism and attachment to political freedom— all which 
conditions were wanting to India. When we reflect that Brahmanism has* absorbed* 
and assimilated every other religious agency that has crossed its path, has attenuated' 
every national or tribal hero into a saint of the Brahmanical pattern, abolished or 
brought into discredit every form of marriage rite except what waa recommended by 
its own teaching, distorted, every legend derived from the heroic age into conformity 
with Brahmanic precepts,, supplanted almost every indigenous- language with its own* 
Sanskrit or Hindi, we cannot be surprised, but should rather as an inevita- 
ble consequence, that it has impressed its own image upon industry also, and that 
every other class in the community possessing some hereditary function should have 
attempted to shape itself upon, the same pattern as- that fnrnished by the Brahman* 

Digitized by 


174. In theory -then, at we have seen, a caste is a tradW*oiu<>n, seeking te» 
secure its own monopoly against all outsiders and denying the right of membership 
to any one not bora within its own ranks from both parent*. Bat theories affecting 
such a complex organism as that of human society ate seldom or never completely 
carried out* The rule about marriage was very simple and could be easily enforced ^ 
for one caste could easily refuse to give its- daughters in marriage to men of other castes,, 
and this practice has beec faithfully adhered to, But hunger is above all rules ; and it 
was quite impossible for any one caste to prevent the members- of any other from en- 
croaching upon it&own function, if sufficient inducement existed. Thus while the caste 
system 'ibm remained very complete and perfect so far as marriage is concerned, k has* 
become,, and was from the first,, far less efficient in regard to- function. Eron the 
Brahman has not been able to secure the monopoly ef his- own function, the priest- 
hood, against all outsiders y for the Goshayen ae temple-priest of Shiva, and the 
Bairagi as temple-priest of Vishnu, have poached, and are still poaching, very 
largely in his preserves (see paras 14ft, 148)* If the most powerful of castes could 
not maintain the monopoly of its own most specialised and jealously guarded func- 
tion—the priesthood— it is not likely that the less powerful castes,, exercising purely* 
secular and less specialized functions, such 8* those of landlordism^ agriculture,, 
•attle-graaing, &>., wonld be able to shut their doors effectually against all encroach- 
ments from without. Thus in all the industries or castes " connected with, the land" 
(as I have described them in my tables of classification?, from the hunter and fish- 
erman to the landlord, there is, and has long been, a constant intermixture of func- 
tion which, owing to the inevitable spread of cultivation and other causes, goes on 
increasing every century. The castes whose functions have been least intermixed} 
are the u artisan castes,' 9 upper and lower, and the <c serving or professional," because in 
these a greater degree of skill is necessary and such skill cannot easily be acquired 
except under a father's guidance. In these, then, the industries have continued 
for the most part hereditary up to the present day, and would have remained so- 
gone the less, if no attempts had been made to secure them against outsiders by the 
rules of caste or trade-unionism. Commerce haa always been open to all comers; and 
as no speciality of skill is required for buying and selling, the "trading " castes are 
perpetually receiving new accessions from without, and ruined traders are perpetu- 
ally entering into occupations connected with the land. 

17!>. Caste, then, though the primary object of its existence was to impose res- 
trictions on trade, craft, or function, has interfered but little with the natural course 
ef industrial activity, and it is net on this account that it deserves the reprobation 
which it has received. A people like that of India, which stopped short at the 
agricultural stage, and to which the scientific and commercial spirit of modern 
times was wholly unknown, could not have been expected to make greater progresa 
than it has done in the arts and industries of civilized life j and caste, by insisting 
en the hereditary nature of functions, haa aided, rather than retarded, the acquisition 
ef artistic skill. Moreover, as several different castes or marriage unions have 
assumed the same function or calling, the element of competition, so necessary to 
industrial activity, has not been wanting. The evil which caste has inflicted upon the 
people of this country is physical and moral rather than industrial; physical, because 
the race has deteriorated through the restrictions placed on the natural freedom of 
marriage ; moral, because the chief motive to union, confidence, and respect between 
the different classes of the community has been destroyed. Society, instead of being 
constituted as one organic whole, is divided against itself by inorganic sections like 
geological strata. The sense of insecurity thus engendered could not but lead to a 
loss of independence and courage in the characters of individuals. For a man soon 
ceases to rely on himself if he thinks that no reliance is to be placed on the good 
will and fair dealing of those around him, and that everything which he may say or 
do is liable to be suspected or misconstrued. Thus the two greatest defects in the 
Indian character — a want of reliance on one's self and a want of confidence in 

Digitized by 


{ 104 ) 

^others — have sprnng from a common source — the terror-striking influence of oastet. 
The caste arrogance of the Brahman, which first sent these evil spirits abroad, has 
-corrupted the whole nation and descended to the very lowest strata of the popnlation. 
A scavenger, the lowest of castes, is as proud of his birthright, and almost more 
punctilious about its rules, than a Chattri <or a Kayasth. His arrogance has even led 
to a disruption within his own fraternity ,: for some Bhangis will no longer eat what 
comes from their master's table, and these men have formed themselves into a distinct 
•sub-caste on that basis. Every Bhangi despises the Cham&r because he eats carrion ; 
while the Chamar despises the Bhangi because he earns his living by removing refuses. 
The lower castes, having no valuable monopoly to defend, have set up the most frivo- 
lous distinctions as grounds for despising their neighbours. Not only has caste 
demoralized society at large, but it is a constant source of oppression within its own 
particular ranks. If a man, in some incautious moment, happens to do something 
forbidden by caste rules, some spy at once swoops down upon him like a bird of prey 
.and summons a panchayat or -caste council, by whose decree he is compelled to per- 
form penance, on pain of expulsion from the caste, by giving * banquet to the 
biradari or "brotherhood" falsely so called. Caste is therefore an instrument both 
of widespread disunion abroad and of the meanest tyranny at home ; and the latter 
of these evils has intensified the want of courage and self-reliance, to which we have 
lately alluded as being one of the greatest defects in the Indian character* 

It is of course quite foreign to the purposes of this paper to distuss either the 
merits or the failings of the natives of this country. The latter have been alluded to 
in this place, only because they appear to me to be the natural outcome of the institution 
of caste — an institution for which no individual is responsible, but for which the 
character of many individual men has been made to suffer. The existence of sucb 
defects does not at all detract from the fidelity and high sense of gratitude observ- 
able in the native character. 

1 76. The other question to be examined was-^In what relation does carte stand 
to the earlier organization of the tribe which it superseded ? 

The history of almost every country, if we look to the earliest stage of its career, 
takes us back to a period when the population was divided into tribes and tribes 
sub-divided into clans. The bond, which first brought the different clans together 
and formed them into a single tribe, has been different at different times and places. 
It may have been some particular deity whom the various clans agreed to venerate as 
their common protector, or some particular river or forest which they regarded as 
the common cradle of their race, or* some fortress which they used as a rallying-point 
at times of attack from without. Or clans may have been drawn together by some 
peculiarities of custom, apparel, or totem. The most usual bond, however, has been 
that of attachment to some victorious chief, who welded the clans together into a com- 
pact brotherhood for purposes of war or defence "against neighbouring tribes or con- 
federacies. The man who thus founds a tribe comes in after years to be regarded 
as its ancestor, and the idea of kinship or a common ancestry thus formed is one of 
the strongest of ties. A tribe, then, is simply a collection of clans called by a common 
name, and bound together by the possession of a common deity, or a common domain, 
or a common totem, or a common ancestry, or common customs, traditions, pursuits, 
&c, or by two or more of these bonds combined. 

177. What we are most interested in observing in this place is that the bond 
of union which first drew the clans together and fanned them ipto a single tribe was 
not unity of craft or occupation. In fact, as each tribe aspired to be a self-sufficing 
body, independent of all other tribes for the tools, weapons, food, and other commodi- 
ties necessary to its existence, diversity, rather than unity, of craft among the compo^ 
nent clans was the thing desired, for the peaceful interchange of commodities among 
i*ibes, whose normal relation towards each other was ,one of war, was of course the 

Digitized by 

Google \ 

( 105 ) 

last thing to be expected. For example, it has been said of the Aht tribe in Vancou- 
ver's Island, " that, though living only a few miles apart, the tribes (clans) practise 
different arts. One is skilled in shaping canoes, another in painting boards for orna- 
mental work, or making ornaments for the person, or instruments for hunting and 
fishing. Individuals as a rule keep to the arts for which their tribe (clan) has some 
repute, and do not care to acquire those arts in which other tribes (clans) excel. 
There seems to be among all the tribes (clans) a sort of recognized monopoly in certain 
articles produced or that have long been manufactured in their own district " (Scenes 
and Studies of Savage Life, by G. M. Sproat, page 19). Sometimes the clans 
most skilled in handicrafts fall into the position of serfs or inferiors to some dominant 
clan, which, being more skilled than the rest in fighting, undertakes the entire respon- 
sibility of defending or enlarging the tribal domain. This is true, for example, of the 
Kandhs or " mountaineers " inhabiting the steep and forest-covered ranges which 
rise from the Orissa coast. Here the Kandh proper " engages only in husbandry and 
war," while the inferior clans supply families of hereditary weavers, blacksmiths, 
potters, herdsmen, and distillers*' (Hunter's Indian Empire, page 77). Almost all 
the indigenous tribes which once covered the plains of Hindustan have long been 
dispersed into castes; but such few remnants as have survived display a greater 
versatility of genius than the castes or functional unions which have succeeded to 
their places. The Kanjar tribe, for example, is skilled in skinning off hides, making 
drums, collecting drugs from forest herbs, trapping the jackal, shooting birds with 
the pellet bow, manufacturing mill-stones, plaiting osiers into mats and baskets, 
twisting cotton into thread, digging for kaslcas roots, manufacturing the weaver's 
brush, Ac. : such versatility will not be found among the lower castes of the Hindu 
community. Indeed, the industries practised by Kanjars are the very industries 
upon which several of the lower castes, such as Cbamfir, Behna, Dharkar, Bhuna, &c, 
have been formed. 

178. In reply, then, to the question which has been asked concerning the relation 
of tribe to caste, I should say that caste acts as the solvent of tribes. The formation 
of a caste implies that clans or fragments of clans, possessing some craft or occupation 
in common, but belonging to different tribes, have been seceding from their respective 
tribes and forming themselves into a new group united by the common craft or industry. 
The new group thus formed is a caste or trade-union. There is no bond of sym- 
pathy in human society that draws and binds men more closely together than that of 
a common industry or (which comes to the same thing in the end) a common interest. 
It is customary in our own language to speak of men, who are of the same cloth or 
practice the same calling, as "brethren "; and btradari or brotherhood is one of the 
terms used by the natives of India to express the caste relationship. The process, by 
which tribes are broken up and trade-unions or castes formed out of the fragments, is 
much facilitated, and in fact almost invariably preceded, by conquest ending in the 
amalgamation of tribes into a nation under some central government. So long as 
tribes are at war with each other, and each tribe is thrown upon its own resources for 
the supply of all its wants, diversity of industries or crafts among its members is 
necessary to its existence. But when tribes hitherto at war have lost their inde- 
pendence and become joint subjects of some larger kingdom or empire, commodities 
can be freely interchanged, and industries are thrown together to an extent which 
was impossible in the previous state of things. The tribe, having thus served its 
end and become useless, begins now to crumble away and disappear, and groups of 
men associated in hereditary professions, trades, or crafts succeed to its place. The 
ceremony or contract which completes the detachment of olans or families of men 
from their ancestral tribe, and renders all return to it or admission into any other 
association impossible, is marriage. When two or more fragments, drawn from differ- 
ent tribes, have been thus cemented by marriage into a single group, and when 
marriage within the group has been made a condition of membership (vide para. 169), 
a caste has been completely formed. This crumbling away of tribes into their com- 
ponent elements, followed by the regrouping of men by some hereditary craft or 


Digitized by 


( 106 ) 

industry, whether secular or religious, is a process so simple and natural, and in its 
season so useful and beneficial, that there are few, if any, nations in the world in 
which it has been wholly unknown. What stereotyped the system in India, and. 
prolonged it to a season far beyond its capacities for usefulness, was (as *e have 
shown) the caste arrogance of the Brahman, It is (this which by the contagion of 
its own example has infected not only the upper, but the middle and even the lowest 
classes of the community who have no craft or industry worth inheriting or defending, 
and has thus multiplied the number of castes on Indian soil to an extent unparalleled 
in human history. 

179. In order to illustrate the process by which tribes become gradually dispersed 
into castes, after they (the tribes) have been first amalgamated into a nation as parts 
of a larger whole, we will select a few examples from the tribes and castes existing at 
this day in Upper India, 

The Kols, with similar aboriginal tribes, such as Bhars, Cherus, Sioris, Doms, Ac., 
were once occupants of a large portion of the plain of Hindustan. Some sixty- 
three thousand Kols, according to the census of 1881, may still be found in the dis- 
tricts south of the Ganges, where their condition is but little, if at all, advanced 
above that in which their forefathers were living some three thousand or four thou- 
sand years ago. A large number more, living in a similar status, may be found in 
the hills and forests of Central India. Those who remained in Hindustan, north of the 
Ganges, have been dispersed, partly into a semi-savage tribe called Musarhas, who 
are only now beginning to settle on the outskirts of Hindu villages as watchmen and 
pig-rearers— partly into cattle-grazers, amongst whom they appear as a section called 
" Kor" of the large pastoral caste of Ahir — partly into agriculturists of the caste of 
Kachhi, which caste is in some places called Koeri, in others Murao, and in others 
Sani, each of these names having been derived from tribes which have helped to make 
it up. Here, then, we have the three successive stages of industry— hunting, nomad- 
ism, and agriculture, all represented by portions of the single tribe of Kol. Other 
portions of the tribe have gone into the various handicrafts. Some (there is rea- 
son to believe) took to iron work and have contributed to the Caste of Lohar. Others 
took to weaving ; and the weaver oaste is better known to this day by the tribal name 
of Kori than by the functional names of Bunkar or Joria. Others took to hide-dress- 
ing and tanning, and appear as a section called Kori in the caste of Chamar. 

180. Another example may be taken from the tribe of Dom. The sites of 
ruined forts and the names and traditions attaching to them show that Doms, Dom- 
katars, Domras, or Donwars (for all these names occur) were once a powerful race 
in Hindustan, especially in the districts north of the river Gogra. The Chattri clan 
of Don war, many of whom still reside in the Gorakhpur and neighbouring dis- 
tricts and who profess to belong to the venerable Bharadwaj gotta or order, bear 
living testimony to the power once held by this widely extended tribe, But While 
some few succeeded in acquiring territorial sway and raised themselves into the ranks 
of the Chattri caste by attaching bards and Brahmans to their persons and learn- 
ing the ways and manners of orthodox Hindus, the rest can be traced only in castes 
standing much below them in rank and function. Some have gone into the Bhangi 
or sweeper caste ; others into the Dharkar or basket-making caste ; others have 
contributed to the Turha or fishing caste; others to the Dhobi or washerman caste ; 
others to the Dbanuk or bow-making caste, from which the Muhammadan caste 
of Kamangar has sprung ; others to the Muhammadan caste of musicians known 
by the name of Dom Mirasi. The remnant of the aborignal tribe, out of which so 
many castes or sections, from the warrior to the sweeper, have descended, is a coarse, 
repulsive savage, air eater of dogs, a burner of corpses, and executioner of the living. 
But in Kuaiaun, where some of the better portions of the original tribe have survived, 
Doms are the chief artisan class and are employed as carpenters, masons, and the 
like, which would lead one to infer that the same tribe contributed many of its 
families in ancient times to the corresponding castes ia Hindustan. 

Digitized by 


( 107 ) 

181. Another and still more striking example may be taken from the tribe of 
Gaundj which is pronounced and spelt in English as Gaur, one of the most conspicu- 
ous names in the caste system of Upper India. The original tribe is no longer to be 
found, so far atf I know, in its purely sarage state, as in the case of the Kol and Dom. 
But we have a near approximation to the aborignal savage in the large fishing 
tribe (for we can scarcely call it a caste, though it has outwardly assumed 
the status of one), which under the name of Gaur or Gaund is still very numer- 
ous along the river tracts of Oudh and the eastern districts of the North- West- 
ern Provinces. Like most of the other aboriginal races, the Gaurs were once a 
widely dominant tribe, and have given their name or some variation of it, not only to 
•many of the villages of Northern India, bat to the large district, which is spelt and 
pronounced in English as " Gonda," but which the natives themselves pronounce (with 
a nasal accent) as " Gaur a." When it is remembered that the highest and noblest 
pursuit of these aboriginal tribes, in times before they were denationalized by Brahmani- 
cal usages and broken up into castes after the Brahmanical model, was that of war, 
we ought to feel no surprise at the feet, that a large and powerful clan of Chattris 
or Rajputs, called by the name of Gaur, exists at the present day. Between the Gaur 
Chattri and the Gaur fisherman there are many other Gaurs scattered among the 
intermediate castes. For instance there is a Gaur sub-caste of Kayasths or writers ; a 
Gaur sub-division of Tag&s — a community of laud-holders ranking a few degrees below 
Chattris proper ; a Gaur sub-division in the caste of Darzi or tailor ; another in that 
of Halwai or confectioner ; and another in that of Barhai or carpenter. In paragraph 
64 1 had already given expression to the conjecture, that the caste of carpenter took 
its origin out of tribes to whom fishing was the chief occupation; and my reason for 
supposing this was that the first fishermen must have manufactured their own boats, 
—those single-logged canoes which are so commonly found among backward races 
in other parts of the world, and are still very common in India. This conjectnre is 
now unexpectedly confirmed by finding the name "Gaur" occurring among the sub- 
castes of Barhai or carpenter, while the Gaur tribe as distinct from the carpenter caste 
is still addicted to boating and fishing. 

We have thus shown how largely the aboriginal Gaur tribe is represented among 
the various castes or functional groups of Upper India at the present day. But there 
is one other caste more important than any yet named, in which the name of Gaur 
holds a conspicuous place, and that is the Brahman* To one who is accustomed 
to look upon Brahmans as being in the main pure-blooded descendants of the 
ancient Aryan invader, the appearance of such a name in such company is at first 
sight rather startling. But as the caste of Brahman is known to have been closely 
allied from the very first to that of Kshatriya or Chattri, and as the name of Gaur, 
together with those of many other aboriginal tribes or backward castes (of which a 
long list could be given if necessary), appears among the sub-divisions of the modern 
Chattri, no surprise need be felt. My own conjecture is that the Gaur sub-caste 
of Brahman is made up of three elements, which were eventually amalgamated 
into one by intermarriage : ( I) those who were nearly related by blood and by social 
position to Chattris of the same name and tribe, but who assumed priestly functions 
in the place of kingly ones, after the example handed down from kings of the Yedic 
age ; (2) Brahmans of other than Gaur origin, who were domiciled in the Gaur 
country under the protection of the Gaur kings and landlords, and who therefore 
received the name of Gaur ; (3) priests of the aboriginal Gaur tribe itself whom 
Brahmans of the two kinds just named admitted into their own caste by investing 
them with the sacred thread. The motive which drew these three elements to- 
gether was that of a common function and common interest. The bond which 
cemented the union and consolidated them into a disjtiqct social unit was marriagp. 
Intermarriage once established, the Gaur sub-caste of Brahman was completely 
formed. The caste once formed retained its generic name of Gaur, but sub-divided 
itself into gotras or orders similar to those assumed by the other Brahmanical 

Digitized by 


( 108 ) 

sub-castes. All tbis of course is only a conjecture ; but the process described is in 
accordance with facts, which are known to have occurred in other instances, and 
with the principles on which castes in general are believed to have been estab- 
lished. This theory seems at least to be more probable than that which derives the 
Gaur Brahman of Upper India from the Gaur Brahman of Bengal. The truth 
probably is that the Bengali Gaur was a colony of Hindustani Gaurs who were planted 
in the Gangetic delta about a thousand years ago by King Adhishwara. 

182. We have seen, then, that a caste is formed out of fragments drawn from 
various different tribes ; that the motive which brings such fragments together is com- 
munity of function and interests ; and that the bond which finally consolidates them m 
into a distinct social unit or caste is the establishment, by mutual consent, of the 
right of intermarriage. We may now exemplify this process once more by showing 
how new castes have been formed, not out of the fragments of tribes, but oat of 
fragments of castes which were themselves previously formed out of the fragments 
of tribes. The process in this latter case would naturally be the same as that in the 

183. It has been usual to speak of Indian castes as being " fissiparous, " like 
those animals and plants in the natural world which are continually throwing off 
fragments or sections from the main body, each fragment acquiring after separation 
a complete individual life. If this metaphor is intended to express a scientific fact, 
I believe it to be entirely misleading. A plant or animal of the kind in question 
acquires its separate life from a single fragment detached from the parental stock. 
I question, however, whether any body of men in India can ever become so entirely 
detached from their hereditary caste as to form a new caste of their own, until they 
have become amalgamated by marriage with some other body of men drawn from 
some other caste. When two or more fragments have thus been united by marriage 
into a single group, a new caste has been formed, and return to the ancestral caste 
on either side is then impossible. 

The Ehattri or banking caste, for example, has in the main sprung (as its name 
implies) out of the great generic caste of Eshatriya or landlord ; and it is 
easy to see how the possession of wealth, implied in the term landlord, should have 
led certain members of this community to devote their money to some purpoje other 
than that of acquiring new land or living in idleness on the acquisition. But the mere 
change of function from landlordism to banking could not by itself have detached 
them from the caste of their forefathers, or prevented the possiblity of return to 
it ; for any man, whatever his caste may be, can take to trading or banking, if he has 
the capital, and without forfeiting his caste by so doing. Moreover, the caste of 
Eshatriya is one to which any man would be proud to belong, and which no one 
would desire to leave without necessity. I hold, then, that the cause which detached 
the Ehattri from the Ohattri or Eshatriya and shut out all possibility of return to the 
anoestral caste was the establishment of a marriage union between fragments or clans 
drawn from several different sub-castes of Eshatriyas, between whom no connubial 
rights had hitherto existed, or from sub-castes of Eshatriyas mixed with those of 
Brahmans. There is much reason to believe that Brahmans as well as Eshatriyas 
have contributed to form the new caste of Ehattri ; for Brahmans of the Sarasvat sub- 
caste will to this day eat food cooked by Ehattris, but no Brahman of any sub-caste will 
eat food cooked by a Eshatriya. Several other circumstances might be named which 
point to the same conclusion. 

184. To take another example, the new caste of Bohra. This is a banking and 
money-lending caste, whose name is derived from beohdr, which signifies " business." 
They are undoubtedly an offshoot from Brahmans ; for the tradition of their origin 
is so recent that many persons still speak of them as Brahmans, though they have in 
fact become quite distinct. Now the word " Brahman," like the word Eshatriya, is 
the generic title for a large number of separate sub-castes ; and what distinguishes one 

Digitized by 


( 109 ) 

sub-caste from another, and entitles it to be called a separate unit, is the fact that 
no one is permitted to many outside this circle. If we suppose, then, that the Bohra 
caste was made up of fragments drawn from several sub-castes of Brahmans, between 
whom no right of inter-marriage had previously existed, the detachment of the Bohra 
from the parental stem of Brahman is explained. The establishment of connubial 
rights among fragments of sub-castes, between whom no such rights previously exist- 
ed, creates (as the reader already knows) a new caste at once and renders all 
return to the abandoned communities impossible. 

185. The Kayasth or writer caste might be quoted as another example. Origin- 
ally the younger sons or near relatives of Chattri and occasionally of Brahman land- 
lords, by whom they were employed as managers of estates and collectors of rent, they 
became gradually amalgamated by intermarriage, and formed a new group possess- 
ing separate marriage rights of their own and denying such rights to all outsiders. 
According to the M&nava Code (X, 6). the male progeny of Brahman husbands 
and Kshatriya wives, or of Kshatriya husbands and Vasiya wives, occupied a rank 
between the two and were called Kayasth. It was only natural that men so born 
(if Manu's account of their birtb is correct), having no distinct status, either as 
Brahmans, Kshatriyas, or Vaisyas, among the ranks of the twice-born classes, should 
form a caste or trade-union of their own, the several constituents of which were 
cemented into a single group by the right of intermarriage, 

186. It is the incapacity of fragments drawn from various different castes 
to become amalgamated by marriage into a single brotherhood, which prevents the 
partially developed castes from becoming completely formed. In illustration of this 
we will produce two examples— the Gadahla and the Sangtar&sh. The Gadhala, if 
such a caste existed, would be a caste of ass-breeders; and as such it would take its 
place with the Rewari or camel-breeder, the P&si or pig-rearer, the Khatik or fowl- 
breeder, the Gadariya or sheep-grazer, and the Ahir or cattle-grazer. But the caste 
of Gadahla cannot come to the birth, because the fragments of the various castes 
by whom the industry of ass-rearing is practised (vis., the Bhangi or sweeper, the 
Eumhar or potter, and the Dhobi or washerman) are as proud of their ignoble castes 
as if they were Brahmans, and are therefore too conceited to form a new marriage 
union. The 8angtar4sh or stone-cutter has been described already in para. 62. At 
the bottom of the scale of stone-cutters stands the savage Kanjar, who makes and 
mends millstones. Next to him stands the Khatik, who practises the same craft Next to 
the Khatik stands the Ahir, who cuts stones from the quarries and prepares them for 
the masons. Next to the Ahir, but very distinct from him in every respect, stands the 
Gokain, who, though properly an engraver in wood, is not less skilful in carving stone. 
Next to these stands the Goshain, a priestly caste, some of the members of which 
make idols and other symbols of stone. Above the Goshain stands the Brahman, who 
shares with the Goshain in the art of image-making. In all this list, it would be 
difficult to find any two fragments sufficiently akin to consent to becoming members 
of a new marriage union and thus forming a new caste ; and so the caste cannot at 
present be said to exist. 

187. Although caste has, as we have shown, acted as the solvent of tribes by 
dispersing the component clans or fragments of clans into different directions, and then 
gathering them up again into new groups, drawn together as by a loadstone into the 
brotherhood of a common industry, yet the organization of the tribe or primary group 
appears to have been the model, upon which that of the caste or secondary group was 
constituted. If castes were formed out of tribes, it was only natural that the child 
should bear some likeness to its parent. The resemblance has shewn itself in two 
different ways: (1) in the rules for the regulation of marriage within the ranks of the 
caste itself ; (2) in the prohibition of intercourse in the matter of eating and drinking 
with outside communities. A few words on each of these points will be necessary 
before closing the subject. 


Digitized by 


( no ) 

188. First, then, as to the rales in regard to marriage. Every caste or sob- 
caste is sub-divided into sections or smaller groups, which, so far as the upper castes 
are concerned, are called gotras. The number of gotraa to any given caste may vary ; 
but whatever the number may be, the caste as a marriage union is made up of the 
sum of these gotras. On the one hand, no member of the caste is allowed to marry 
outside the circle of this union. On the other hand, no one is allowed to take a wife 
from the same gotra as his own. Among all the upper castes this is a clearcut rule, 
to which no exception exists. The explanation given by Brahmans as to the reasons 
why a man is thus prohibited from taking a wife of his own gotra is that every gotra 
is descended from some Rishi or sage of the Vedic age ; and consequently alt persons 
born within its ranks are in the relation of brother and sister. Marriage, therefore, 
within the same gotra would be incest. Now this gotra system of Indian castes 
appears to bear a close analogy to the clan system of tribes. A tribe of the ancient 
type, like the caste of modern India, was a marriage union, that is, a wider circle 
within which a man was compelled to marry, but containing within its circumference 
a number of clans or smaller circles within which marriage was disallowed. The analogy 
however, though close, is not quite exact. For in the case of the caste it is only the 
gotra, but not the caste or union of gotras, which claims descent from a common ancestor ; 
whereas in the case of the tribe, not only does each clan sometimes set up an ancestor 
of its own, but the tribe or union of clans sets up a more ancient ancestor, who is himself 
supposed to have been the father or forefather of the several minor ancestors. The long 
and sanguinary dispute which the plebeians waged against the patricians of Borne, one 
of whom demanded and the other refused the right of connubium or inter-marriage, and 
the reluctance with which the right was at last conceded, show how very difficult it 
was, even under a republican Government, to break through the line of the outer or 
tribal circle. It is probable, in the absence of any evidence to the contrary, that evety 
Aryan tribe which migrated into India some four thousand years ago had a clan sys- 
tem of its own, similar to that which has prevailed so widely elsewhere ; and a con- 
siderable number of names which appear to be* tribal occur incidentally in the Yedic 
hymns. From what we know, too, of the indigenous tribes of India at the present 
day. — those at least whose original character has not yet been defaced through con- 
version to Hinduism— it is probable that a similar system prevailed among the 
ancient aboriginal tribes, with whom the invading Aryans were brought into contact. 
If, then, the modern Hindu has sprung, as I hold that he has, out of the amalgamation 
of the two races, and if caste has sprung out of the dispersion of tribes, it is no wonder 
that the caste should have organized its own constitution upon the same model as 
that of the tribes out of which it arose. 

A gotra in the strict Brahmanical sense is an exogamous group, the members of 
which are supposed to be descended from a common Yedic saint. As we have stated 
already, such groups are only to be found among the upper or " twice-born " castes, 
that isf, among those which have become completely Brahmanized. Among Chattris, 
but not among Brahmans or Eayasths or other high castes, there are clans of the 
strictly tribal type as well as gotras of the Brahmanical type ; and a Chattri in choos- 
ing a wife decides for himself what are the limitations to his choice as regards the clan, 
but leaves it to the Brahman to decide what limitations exist in regard to the gotra. 
Among 'the lower castes the rule of exogamy is worked in various different ways. 
Sometimes we find a caste made up of clans of the tribal type, that is, clans which 
have retained their identity as such after having detached themselves from the 
parent tribe in order to enter the caste or functional brotherhood. Sometimes we 
find a caste broken up into small exogamous groups, which are known in the caste 
itself as gots, but are not the same as Brahmanical gotraa sprung from a Yedic saint. 
In castes where no clan or got or gotra can be traced the rule of exogamy is always 
applied to the ghardna or circle of blood relationship, on the male side, traced back 
to the fourth or fifth generation. The gharana is of course a much smaller circle of 
prohibited degrees than the modern gotra or the ancient clan. 

Digitized by 


( 111 ) 

The rule of exogamy is found in foil force at the present day among several of 
the aboriginal tribes of this country, snch as the Santals and Kandhs of Central 
India, the Tharus, Nats, Kanjara, and Doms of Upper India, and the Bhils, 
Grassias, and Minas of Bajputana. The clan may not always be of the most 
perfect type— w., that which claims descent from a common ancestor ; bnt the 
rnle of exogamy is strictly enforced. Speaking of the Kandhs, Dr. * Hunter re- 
marks :— " The clan consists of a number of families sprung from a common father, 
and the tribe is made up in like manner from a number of clans who claim descent 
from the same ancestor" (Indian Empire, page 76). Here the smaller group, which 
he calls a clan, and the larger one, which he calls a tribe, alike coincide with the 
Brahmanical conception of gotra — a body of men related by blood on the male side 
through desoent from a common father or forefather. Further on he says regarding 
the same people, " marriage between relations, or even within the same tribe, is for- 
bidden " (page 77). The same writer, speaking of the Santals, says — "The Santals 
know not the cruel distinctions of Hindu castes, but trace their tribes, usually fixed 
at seven, to the seven sons of the first parent :" and of their marriages he adds r— 
" The Santal must take his wife not from his own tribe, but from one of the six 
others " (page 73). If we substitute " tribe or clan" for " gotra," the organization 
of the Santal and Eandh communities gives a very fair representation of that of an 
Indian caste formed on the Brahmanical model. 

189. The other point, in which the discipline of the caste resembles that of the 
tribe, consists in the rule by which intercourse in the matter of eating and drink- 
ing with outside communities is forbidden on pain of expulsion from the fraternity. 
The range or circle of the convivial union, if I may so eall it, is co-extensive with 
that of the marriage union ; for convivium is the necessary sequel to cinnubium. 

Fire and water, food and drink, which (as every one's experience must tell him) 
constitute the life of the individual body, have been regarded in almost every tribe or 
state, with whose traditions and customs we are acquainted, as symbols of the collec- 
tive life of the tribe, state, or nation. " So strong is the bond of race/' says Dr. 
Hunter writing of the Santals, page 73, " that expulsion from the tribe was the only 
Santal punishment A heinous criminal was cut off from fire and water in the village 
and sent forth into the jungle. Minor offences were forgiven upon a public reconcilia- 
tion with the tribe, to effect which the guilty one had to provide a feast with much 
rice-beer for his clansmen." We have only to substitute "caste" for "tribe," and 
the above passage gives as faithful a picture as we could wish of the internal discipline 
of an Indian caste. The symbol of caste membership, or of re-admission into caste when 
membership has been suspended for some offence against caste rules, is a common 
meal cooked at the same fire. Expulsion from caste means expulsion " from fire and 
water." A man who eats food cooked in strange fire and strange water breaks 
the tie tliat binds him to his own fraternity and becomes an outcaste. In the 
Roman commonwealth and in all the Greek States, whose institutions have been 
recorded, the sentence of banishment pronounced against a guilty citizen was expulsion 
from fire and water, and thus between the status of the banished citizen and that of 
an Indian outcaste therfe was no practical difference. In ancient Lacedsemonia the 
Spartiatoe, or citizens of Sparta proper, constituted a distinct marriage union or caste ; 
and the symbolism of caste membership was preserved by the institution of public 
meals (Syssitia). In a large number of communities both of the old and the new 
worlds a public fire was kept constantly burning on a central hearth or altar to 
symbolize the collective life of the horde, confederacy, or state. Among those tribes, 
of Upper India which are still outside the pale of caste, and therefore form no part of 
the Hindu community proper (suoh as Th&rus, Eols, Kanjars, Nats, Doms, &c), the 
rule against eating and drinking with outside tribes » strictly observed. The same 
may be said of tlpose tribes " allied to the hunting or fishing states," who stand in the 
very lowest stratum of Indian castes. A Banjara once informed me that his tribe 
will not even sp^ak, much less eat or drink, with Kanjars. The three great tribes who 

Digitized by 


( 118 ) 

founded the Roman commonwealth, the Ramnenses, Titiei, and Luceres, were at per- 
petual war, till they agreed to mingle their respective fires in the common hearth or 
altar of Vesta and (has constitute a single state. 

190. As life is made op of food and drink combined, and not of food only or of 
drink only,, so in the Ales of caste it is the use of fire and water combined, not the 
use of fire only or of water only, which renders food cooked by one caste unfit for 
consumption by another. Thus any man, whatever his caste may be, can eat parched 
grain (chabena>, that is, grain fried without water. ' Had no such qualification exist- 
ed, no such caste as that of the Bharbhunja, who sells parched grain to the 
general public, could have come into existence. Again, any man may eat flour or 
grain sweetmeats cooked in both fire and water, provided the immediate contact with 
fire is shut off with butter. If the butter qualification had not been thought of, the 
caste of Halwdi (baker and confectioner) could never have sepn the light. The 
Brahman has asserted his superiority over all other castes by making his case excep- 
tional in these culinary matters. On the one hand, whatever he cooks, in fire and water 
both, is privileged so that no one of any other caste can lose caste by eating it. On 
the other hand, his own purity so far exceeds that of other castes that he may not eat 
either the parched grain prepared by the Bhunja or the buttered bread or grain sweet- 
meat prepared by the Halwdi. Considering the immense amount of fass attached by 
all the castes to questions of eating and drinking, the exceptional position, which the 
Brahman has acquired in these matters, is alone sufficient proof of the arrogance of 
his pretensions and of the success with which he has asserted them. 

191. The right of convivium or eating together follows, as we have shown, that 
of conntrhium or intermarriage as closely as an effect follows its cause. But woe to 
the man who inverts the order, that is, who takes food cooked at a strange fire, food 
cooked by some outsider, with whom he does not stand in the marriageable relation, 
and who therefore belongs to a different caste. Such an act, wilfully or knowingly 
committed, entails dismissal from the fraternity, which in the case of the lower castes 
can be redeemed by a banquet, but in the case of the upper ones is absolutely beyond 
remedy. The condition of an outcasts is almost indescirable. He is completely cut 
off from society ; for not only is he scouted by his own brotherhood, but no other caste 
will take him in or make him one of themselves. There are but three sources open to a 
man in such a case. He can either go away with his family to some distant place 
where no one knows him, and give himself out to be of such or such a caste, and thus 
regain some kind of social footing. Or he can renounce the world, attach himself 
to some religious order, and become a Qoshain or some other kind of celibate, leaving 
hia wife and children to be cared for by their relations or by his own : (for his own 
loss of caste does not entail the same infliction on *them, unless they too have commit- 
ted the same offence). Or lastly, he can change his religion altogether and become 
a Musalman or a Christian. I know of one case pf a Kayasth who lost caste by 
eating food which had been cooked at a strange fire. v Every conceivable attempt was 
made by the offender to recover his lost status, but a* 1 ! his offers of penance were 
indignantly rejected. The offence was unpardonable. At ;last, in utter despair, he 
turned Muhammadan by undergoing the rite of circumcision and repeating the 
confession. But even then his wife and family were detained by their relations 
and not allowed to join him. One night, however, by a> preconcerted plan, she 
fled secretly into a house where her husband had arranged to receive her and 
to have some dinner ready cooked for her at his own fire. Sh<B had scarcely had time 
to enter the house and place a morsel of h\p food between her lips, when her pursuers 
rushed in and claimed her. But when they saw that she w(as chewing something 
that he had cooked at his own fire, they at once gave her up *b lost, knowing that 
that one morsel of food had, by the rules of caste, out her otf for ever from all her 

kith and kin. Such is the efficacy in India of food cooked in a ,«tradge fire. 


192. Assuming, then, that " functional union bounded by a marriage uuion" is 
the minimum difinition of an Indian caste, it is necessary to point out in conclusion 

Digitized by 


{ 113 ) 

that tbe number of castes or marriage unions actually existing in the North- West and 
Oudh is farinexoess of the one hundred and twenty-one names shown in the annexed 
tables. To be more precise, we ought to exclude the nineteen names of group I, repre- 
senting " casteless tribes/' and the two names of group VII, representing tbe " religious 
orders/ 9 as neither of these belong to the list of castes proper. This will leave us a 
list of one hundred different castes. Now what we hare to observe is that each of 
these names is merely a major head descriptive of the general function under which 
a large number of minor heads or separate marriage unions are included. For 
instance, tinder the major head " Barrhai or carpenter" (see table, group III B.) the 
number -of minor heads or marriage unions comes to at least seven, and therefore the 
real number of Barhai castes is not one, but seven. Again, under the major head 
" ftayasth " (see table, group X) there are ten clearly-defined marriage unions, and 
therefore tbe actual number of Eayasth castes is ten and not merely one. Under' 
the head "Chattri" there must "be tit least thirty marriage unions ; for in this 
" elastic fraternity" (as Sir H. Elliot has most justly called it) marriage unions are very 
changeable, varying according to the wealth and status of individual families. Under the 
head * Brahman" there must be at least forty different castes or endogamous groups. 
The ^disunion which the Brahman has sown broadcast among the people by his own 
arrogance has thus recoiled with more than double effect upon his own head ; for 
there is no caste in India which is broken tip into so many hostile and mutually des- 
pising sub-castes as that bearing tbe generic title of Brabman. If it were possible to 
ascertain the precise number of marriage unions existing at the present time, this 
would be an exact measure of the number of castes. I believe that I am well within 
the mark when I say that to every major head named in the annexed tables the aver* 
age number of minor heads or marriage unions is not less than ten. According to 
this, then, the total number of castes at the present time in the North-West and Oudh 
is not one hundred, but one thousand. 

193. The number of castes in India is constantly changing, because <every mar- 
riage union may admit new constituents or expel old ones. The circle of marriage 
unionship is not, nor ever has been, immutable. It might rather be compared to a 
circle, whose centre can change its point, and whose radius is perpetually liable to 
being lengthened or contracted. This process of extension in one direction, followed 
as it usually is by contraction in another, has been at work for the last two thousand 
years at least; and thus, while some castes once prosperous, such as the Kathak and 
the Bhat, are surely but not slowly dying out, and while others, such as the Baidya, 
have died out altogether, other castes, such as the K'Wtri and Eayasth, have been 
rising into importance in therr places. When, therefore, a caste is said to die out, 
this does not mean that the families which belonged to it have died or been extermi- 
nated, but either that the function peculiar to this caste has become useless to the 
outside community, in whioh case the families constituting the caste are gradually 
dispersed into other functional and marriage unions, or if the function is stHl useful 
to the outside world, that some new and more energetic centre of activity, giving rise 
to a new marriage circle and assuming a new title, has supplanted the old one. The 
Baidya, as Baidya, has disappeared, but his descendants are still alive under other 

194. It was necessary to allude to the above facts in order to explain the bulk 
to which the Brahman and Chattri, " the first-born and second-born of castes," have 
grown in the course of the last twenty centuries. According to the figures shown in the 
annexed tables, the one contains 4,690,850, and the other 3,127,207 persons. Consi- 
dering that the function of the Brahman is to act as priest, and that of the Chattri as 
landlord or ruler, such numbers are obviously far in excess of the capabilities of the 
outside community. At least three-fourths of the men who call themselves by such 
names can find no scope for exercising the functions peculiar to their respective orders: 
But rather than change their names and fall into the ranks of any of the outside 
castes, whose functions they have assumed as a means of livelihood, they have clung to 


Digitized by 


( 114 ) 

the aneettral brotherhoods. From the very first both of these castes found that their 
interest lay in extension rather than in contraction. There are or were three kinds of 
landlords, rulers, or warriors, whom the original Kshatriya castes have admitted at 
different times into the ranks of their own fraternity by giving them their sons or 
daughters in marriage t— (1) men bearing the names of aboriginal tribes, who founded 
petty kingdoms and established local dynasties ; (2) men bearing the names of some 
of the inferior castes, who rose into power during the ascendancy of Buddhism or from 
other causes ; and (3) a class of chiefs, whose origin cannot be traced to any particular 
time or place, but who called themselves after some animal supposed to resemble them 
ia character, such as Kachaura (tortoise)* Baghel (tiger), Har£ (wasp), Rikh-bansi 
(bear)> Nag-banst (serpent), Sarpakheria (snake), Singhel (Hon), Bains (buffalo). 
It is no wonder, then, that among the Chattri sub-castes or marriage union* 
now existing we find (as was pointed out in paragraph 188) clans of the tribal type 
co-existing with gotras of the religious type; for the word Kshatriya or Chattri is simply 
a general name which came to bo applied, in virtue of its own signification, to men of any 
tribe, olan, or caste, who have at different times risen into power as local chiefs and 
compelled by force of arm* the obedience of the peasants, herdsmen, and others 
living within reach of their forts. Again, there are or were five kind* of priests whom 
the original Brahman caste has found it necessary to co-opt at different times into the 
marriage union:— (1)* Kshatriya*, of any origin whatever, who voluntarily exchanged 
the function of ruler for that of priest ; (3) priests of the aboriginal tribes, who were 
admitted into the ranks of Brahmanhood as fast as their own tribes become Hinduized ^ 
(3) priests of Buddhism who were admitted into Brahmanhood a* fast as* Buddhism 
itself melted back into the older creed ; (4) priests manufactured by kings out of 
various inferior castes for the occasion, of some immense banquet or religious festival, 
when> a sufficient number of Brahmans oould not be found ready made ; (5)>the ille-* 
gitimate sons o£ the various celibate orders, who assumed the status of river-priests or 
temple-priests at the sacred rivers or shrines, where the monasteries stood. It is pro- 
bable too, for reasons given in paras. 48, 33, and 97, that men of the Bari, M&li, and 
N&pit castes have occasionally crept into the ranks of Brahmanhood.. The castes of 
Brahman and Chattri have been expansive and elastic from the beginning. But while 
new members have been taken iu at different times, whose admission seemed likely 
to extend the influence of the castes which adopted them, many of the older families^ 
and olans > whose retention was of no nse for such a purpose, have olung to the- 
ancestral brotherhood with & tenacity which, pride only can attach to poverty. 

195. If it were possible to compress. into a single paragraphia theory so complex, 
as that which would explain the origin and nature of Indian caste, I should attempt 
to sum it up in some suoh words as the following. A caste is a marriage union,, 
the constituents of which were drawn from various different tribes (or from various, 
other castes similiarly formed), in virtue of some industry, craft or function, either 
secular or religious, which they possessed in common. The internaL discipline, by 
which the conditions of membership in regard to connubial and convivial rights are 
defined and enforced, has been borrowed from the tribal period which preceded the 
period of castss by many centuries, and which was. brought to a close by the amalga- 
mation of tribes into a nation under a common sceptre. The differentia of caste as a 
marriage union consists in some community of function ; while the differentia of tribe 
as a marriage union consisted in a common ancestry, or a common worship, or a com- 
mon totem, or in fact in any kind of common property except that of a common func- 
tion. Long before castes were formed on Indian soil, most of the industrial classes, 
to which they now correspond, had existed for centuries, and as a rule most of the 
industries which they practised were hereditary on the male side of the parentage. 
These hereditary classes were and are simply the ooncrete embodiments of these suc- 
cessive stages of culture which have marked the industrial development of mankind in 
every part of the world. Everywhere (except at least in those countries where he is 
atill a savage), man. has advanced from the stage of hunting and fishing to that of 

Digitized by 


( 115 ) 

aemadism and cattle-grazing, and from nomadism to* agriculture proper. Everywhere 
has the age of metallurgy and of the arts and industries which are coeval with it been 
preceded by a ruder age, when only those arts were known or practised which sufficed 
for the hunting! fishing, and nomad states. Everywhere has the class of ritualistic priests 
and lettered theosephists been preceded by a class of less cultivated worshippers, who 
paid simple offerings of flesh and wine to the personified powers of the visible uni- 
verse without the aid of an hereditary professional priesthood. Everywhere has the 
class of nobles and territorial chieftains been preceded by a humbler class of small 
peasant proprietors, who placed themselves under their protection and paid tribute or 
rent in return. Everywhere has this class of nobles and chieftains sought to ally 
itself with that of the priests or sacerdotal order ; and everywhere has the priestly 
order sought to bring under its control those chiefs and rulers under whos* protection 
it lives. All these classes, then, had been in existence for centuries before any such 
thing as caste was known on Indian soil ; and the only thing that was needed to con- 
vert them into castes, such as they now are, was that the Brahman, who possessed 
the highest of all functions — the priestly— should set the example. This* he did by 
establishing for the first time the rule that no child, either male or female, could in- 
herit the name and status of Brahman, unless he or she was of Brahman parentage on 
loth sides. By the establishment of this rule the principle of marriage unionship was 
superadded to. that of functional unionship; and it was only by the combination of these 
two principles that a caste in the strict sense of the term could or can be formed. The 
Brahman therefore, as the Hindu books inform us, was " the first-born of castes." 
When the example had thus been set by an arrogant and overbearing priesthood, whose 
pretensions it was impossible to. pui down, the other hereditary classes followed in 
regular order downwards, partly in imitation and partly in self-defence. To a 
nation mesmerised by Brahmans and blinded with superstition and ignorance no 
other course was open. Immediately behind the Brahman came the Kshatriya, the 
military chieftain or landlord. He therefore was the " second-born of castes." 
Then followed the bankers or upper trading classes (the Agarwal, Rhattri, &c.) ; the 
scientific musician and singer (Kathak) i the writing or literary class (Kayasth) ; 
the bard or genealogist (Bhat); and the class of inferior nobles (Taga and Bhuinhan, 
who paid no rent to the landed aristocracy. These, then, were the third-born of 
castes. In all communities, such classes must stand rather high in the scale of social 
respectability, since the stages of industry or function* which they uepresent are 
high in proportion ; but in India their rank was more precisely defined than else- 
where by the fact that they made a nearer approaoh than the castes below them to the 
Brahmaaioal ideal of personal dignity and purity. Next in order came those artisan 
classes, who were coeval with the age and art of metallurgy;, the metallurgic classes 
themselves ; the middle trading classes ; the middle agricultural classes, who placed 
themselves under the protection of the Kshatriya and paid him. rent in return (Kurmi, 
Kachhi, Mali, Tamboli) ;. and the middle serving classes, such a* NApit and Baidya, 
who attended to the bodily wants of their equals and superiors. These, then, were the 
fourth-born of castea ; and their rank in the social scale has been determined by the fact 
that their manners and notions are further removed than those of the preceding castes 
from the Brahmanical ideal Next came the inferior artisan classes, those which 
preceded the age and art of metallurgy (Teli, Kumhar, Kalwar, Ac.) ; the partly 
nomad and partly agricultural classes ( J&t, Giijar, Ahir, &o.); the inferior serving 
classes, such as Kah&r ; and the inferior trading classes, such as Bhunja. These, 
then, were the fifth-born of castes, and their mode of life is still further removed 
from the Brahmanical ideal than that of the preceding. The last born, and therefore 
the lowest, of all the classes are those semi-savage communities, partly tribes and partly 
oastes, whose function consists in hunting or fishing, or in acting as butcher for the 
general community, or in rearing swine and fowls, or in discharging the meanest 
domestic services, such as sweeping and washing, or in practising the lowest of human, 
arts, such as basket-making, hide-tanning, &c. Thus throughout the whole series o{ t 
Indian castes a doable test of social precedence has been in active force, the -Industrial, 

Digitized by 


( 116 ) 

and the Brahmanical; and these two have kept pace together almost as evenljr* 
as a pair of horses harnessed to a single carriage. In proportion as the function 
practised by any given caste stands high or low in the scale of industrial develop* 
ment, in the same proportion does the caste itself, impelled by the general tone of 
society by which it is surrounded, approximate more nearly or more remotely to the • 
Brahmanical ideal of life. It is these two criteria combined which have determined 
the relative ranks of the various castes in the Hindu social scale. Outside the caste 
system altogether stand the few and shattered remains of those aboriginal tribes, out 
of which the whole series of castes was fashioned by slow degrees, through the example 
and under the guidance of the Brahmanical priesthood. Had the Brahman never • 
come into existence and had his arrogance proved to be less omnipotent than it did, 
the various industrial classes would never have become stereotyped into castes, and 
the nation would then have been spared a degree of social disunion to which no paral- 
lel can be found in human history. There seems to be no likelihood of caste being • 
banished from Indian soil until Brahmanism itself— the font et origo mali — has died 
a natural death by the rise of the scientific spirit, and the fallacy of its pretensions ■ 
has become an object of general scorn. As soon as the Brahman begins to disappear, 
the rest will follow. 


196. The census report gives us very little information concerning the Jains, 
and that little is given doubtfully. In section IX, on the " Religions of the People/* 
itsays nothing about their tenets, and in section XXIII it telle us nothing about 
their castes. In form III, which shows the statistics of population according to 
religion, the total number of Jains, as distinct from Hindus, is set down as 79,957. 
This is in fact the only thing that the report tells us about them, and even this, as 
the compiler explains in paqe 59 of the Preliminary Dissertation, is probably incor- 
rect : — a The Jains or Saraogis in these provinces are generally regarded as a sect of 
Hindus. Under the orders issued to enumerators by the Government of India, 
the castes and not the sects of Hindus were to be recorded, and consequently many 
Jains were returned as Hindus simply. Much was done by subsequent inquiry to 
correct this ; but I have no doubt the number of this sect is considerably under- 
stated." The italics here given are the compiler's own. But there are some castes 
which are exclusively Jain ; and hence it would not have been opposed to the orders 
of the Government of India to have collected the statistics of population assignable to 
each of tiiese. 

197. Jainism, though a keen rival to Buddhism, which it has regarded all 
along with the most cordial hatred, has yet been always called a sister creed or > 
satellite, and the very resemblances between them afford a natural explanation if 
their hostility. If current traditions are to be trusted, both creeds made theii} 
appearance in the north of India at about the same time (viz., about 550 B.O.) What-\ 
ever the differences may have been among themselves, there was one sentiment at - ■ 
least which they shared in eommon — antagonism to the pretensions and teaching of . 
Brahmans. The word "Jain "means a follower of Jina the Conqueror, just as 

" Baudha " means a follower of Buddha the Enlightened. The word for " follower*' 
in both of these creeds was " Sr&vak," a listener or- disciple, and this word appears 
to have given rise to the name " Saraogi," by which Jains are not unfrequently 
called to this day. In ancient times, the most famous centre of Buddhist and * 
Jainist teaching in Upper India was the city of Srarasti (in the Gonda district, . 
Oudb), the ruins of which are now at last being exhumed : and the name Srivasti 
is probably a contraction of Sravak-sthiti, " the abode of the disciples," and not, as 
Mr. Benett has suggested in the Oudh Gazetteer, a derivative from Savitri, a Vedio > 
name for the Sun-god. The saints and founders of the Jainist creed were called Jinas • 
or conquerors, in virtue of the fact that they had prevailed over the temptations of . 
the world, and could point out the way, the truth, and the life to all who desired . 
a similar emancipation from the coils of mortal existence. They are also called. 


Digitized by 


(,117 ) 

Tirtbankfirs, which simply means " teabher, preceptor" (from tirtham, precept, alid idr,- 
maker or doer), m contradistinction to SrAvak, disciple.. There are said to have beeu 
tWenty-four Jiqas or Tirthankars, spiritual conquerors or spiritual preceptors, who 
appeared on the earth at wide intervals of time apart. The first of the series was 
Rishab, and the last two PaVswa N&th and Mahavir : the names of the twenty-one 
intermediate ones are very seldom heard of. The last, Mahavir or the " Great Hero," 
is said to have died in 526 B.C., and the almost universal tradition recorded in the 
Jainist Scriptures points to this man as the founder of the creed. The original tenets 
of Jainism were decidedly opposed to those of the Brahmans. It taught that there 
was no caste and no deity, and that the slaughtering of animals in sacrifices was 
an impious encroachment on the rights of the sentient creation. It denied • the 
inspiration of the Brahmanical Vedas and set up a new Veda or inspired book of its 
own. It dwelt very strongly on the sanctity of animal life of all kinds, but authorized 
the slaying of a Brahman as the arch-enemy and seducer of mankind. 

198. For several centuries past, however, Jainism has been steadily drifting 
back into the older creed of Brahmanism, from which it at first seceded with so much 
acrimony : and hence, if we look to the present state of Jainism and ignore its begin- 
nings, it is now correct to say that its followers are merely a sect of Hindus. 
Probably, too, the disappearance of the sister creed, Buddhism, from Indian soil was less 
due to open violence than to the silent imperceptible power which Brahmanism has 
everywhere possessed of absorbing or winning back every form of schism or indepen- 
dent thought that has sprung up among the people ; and this is the opinion now gener- 
ally held by scholars. Most, if not all, Jains of the Upper Provinces are not only ' 
content to call themselves Hindus, but would reject with soorn the imputation of being 
Buddhists or of having any affinity with them, though I am told that in Bajputana the 
assimilation of Jains to Hindus is much less complete than elsewhere. The denial 
of the existence of gods or supernatural beings, on which the first Jains so warmly 
insisted, has not prevented their descendants from reverting to a kind of theism or' 
polytheism of the Hindu pattern. The twenty-four Jins or Tirthankars have been 
raised to the rank of deities, and colossal images are erected in their honour at the Jain 
temples, notably in those of the Dekkan. Side by side with these deified mortals are a 
class of goddesses named Sh&sanadevis, who execute their commands, and whose presence 
in such company appears to be merely an adaptation of the doctrine of the Saktis or 
female powers associated with the three great gods of Hinduism. In addition to these, 
many other deitieB are recognised, all of whom have been borrowed one by one from 
the Hindu pantheon. The Jains have even multiplied the gods of the Hindu 
firmament by recognising four Indras or sky gods instead of only one. But they do not 
profess to give any preference to the three great gods of the Hindu Triad— Brahma, 
Vishnu, atod Shiva ; and theoretically all gods borrowed from the Hindus are subordi- 
nate to their own saints — the Jinas or Tirthankars. Even this theory has, however, 
been virtually set aside of late by the modern doctrine now generally accepted by 
Jains, that Rishab or Rishab N&th, the first of all the Jinas, was an avatar or 
incarnation of Vishnu. There is a temple in the south-west of Mewar (as Mr. Ibbetson 
points out in his late Panjfib Census Report), where Rishab is worshipped as a form 
of Vishnu by Jains and Hindus alike : and thus, if we are to look upon the Jains as a 
sect of Hindus, they belong to the Vishnuite rather than to the Shivite school of 
believers. Though the first Jains discarded Brahma, the great being whom Hindus 
have from the earliest times regarded as the supreme power in the universe, yet their 
descendants have put another being in his place, whom they call Jinapati, which being 
interpreted means " the lord of the Jinas or conquerors." The name Varddhaman 
which is sometimes applied to.Mah&vir, the traditional founder of Jainism, savours 
strongly of the Hindu doctrine of Brahma; for it is derived from the root vridh 
to grow or expand, just as the word Brahma is derived from brih, which is only 
another form of the same verbal root. Ail this betokens a very wide departure from 
the original atheism with which the creed set out. Brahmanism ixas in fact been too 


Digitized by 


'( "« ) 

strong' for this, as it has been for every other, form of dissent, which has attempted 
to found a church in India outside ike range of Brahmanic easte and ritual 

199. The tendeney thus displayed by Jains to revert to the ways and notions 
of Hinduism is equally, if not more, marked in social than in religions matters. 
Notwithstanding all the professions once made about the rejection of Brahmanic 
pretensions and the natural equality of all men, they have themselves become broken 
up into castes fashioned after the Brahmanic model. The names of these castes, so 
far as I can learn, are Osw&l, Sri MM, Sri Sri M&l, Sri M&l Pat tan, Bauddhmati, and 
Parwar, six castes in all. These are castes in the strict sense of the term — that is, they 
are trade uuions which deny all right of intermarriage or convivial intercourse with 
outsiders. I learn, too, that in the disposal of property they follow the Hindu law 
of inheritance and call iu Brahmans to decide doubtful points of succession. The 
name " Banddhmati " is evidently of religious origin, and implies that the first 
members of the caste were followers of Buddha, though their descendants are now 
without exception Jains. The remaining five names appear to be all geographical, 
and to indicate that the castes now settled in Hindustan came originally from 
Central or Southern Iudia. Thus the Osw&ls, the most important and influential of all 
the Jain castes, say that they came from Qsia, a town in Jodbpur. The name "Sri 
Mai " is derived from the country M&l or Malwa, and the prefix " Sri " has been 
added merely to denote the good luck or prosperity claimed by the caste. The re- 
duplicated " Sri " in the caste name " Sri Sri Mai" merely indicates that this rival 
fraternity claims a higher degree of prosperity than the other. The " Pattan" at- 
tached to the name, Sri MAI Pattan, implies that a portion of the caste came from 
the south, where " Pattan " is a common name for town or city. This caste, then, 
represents a marriage union between some Jains of Mai or Malwa and certain other 
Jains who came from further south ; and as such it exemplifies the method on which, 
as 1 have attempted to show in this paper, castes in general have been formed. Par- 
war is merely a contraction of Palli-war (the name by which the caste is sometimes 
called), Palli being a town in the Jh&nsi Division. The etymologies here assigned, 
aud the inference to be drawn from them as to the quarter from which the castes 
immigrated into Hindustan, are confirmed by the fact that almost all the distinguished 
shrines and holy places frequented by Jain pilgrims are in Central Or Southern, and not 
in Northern India ; as for example, Mount Abu to the north of Gujarat ; the hill Parswa 
Nath (generally misspelt as Parasnath), on the top of whioh P&rswa, the last but 
one of the Jinas, is said to have died — that is, to have obtained his nirvdn or libera- 
tion from earthly life ; the rock-cut caves at Ellora, Naaeik, and other places ; and the 
huge rock-cut idol of a Jina near Chinrai Pattan, in Mysore. There are many 
remarkable temples, too, in the state of Datia, near Gwalior. The secluded and moun- 
tainous character of these places appears to imply that, though Buddhism had been 
banished from the plains of Hindustan, the Jains were able to hold out in these safe 
retreats until the period of Mubammadan domination had set in. This domination, 
while it checked the progress and arrogance of Brahmanjsm, was tolerant of smalt sects 
or religious associations, from whom no political danger was anticipated ; and under 
the protection thus afiorded, certain Jain communities moved northwards and settled 
in the cities of Hindustan, where we now find them. In physiognomy, so far as I 
have been able to see, the Jain castes of Upper India, especially the Oswal, are fairer 
aud better-featured than the average Chattri or Brahman, and appear to have retained 
a stronger dash of Aryan blood. 

200. The chief, and in Upper India the only, hereditary industry in which the 
Jain castes engage is that of commerce or banking. The six castes named in the pre- 
vious paragraph are pre-eminently trading castes, and would have been grouped undo): 
that heading in my appended statement, if the census report had furnished the names 
and figures. In almost all the large towns from Lthore to Bombay or Calcutta they 
hold a conspicuous place ts traders or bankers ; and the particular aptitude whjch 
they evince for this kind of calling is well in keeping with the prominent part played 

Digitized by 


( 119 ) 

by merchants, bankers, goldsmith*,- Ao, in the legends and chronicles of Bnddhism, 
die sister creed. This* however, does not prevent individuals from taking service 
as writers, clerks, pleaders, accountants, &c., in all which capacities they have become 
the rivals to Kayasths and Brahmans, to whom they are in no respect inferior in 
intellectual quiokness. The wealth which they have accumulated in banking or service 
has enabled some of them to purchase landed estates and set up as landlords, but 
not as warriors, after the manner of Chattri*. In Central India, as for example in 
the Jbansi Division, 1 learn on the authority of an eye-witness that many of them 
work in the fields as tenant farmers or labourers ; and one writer affirms that to the 
Jains of Southern India agriculture is the usual occupation. This proves as conclu- 
sively as any argument that could be adduced the extent to which all castes in In<Jia, 
whatever their hereditary function may be, are being gradually thrown upon agricul- 
ture as a means of livelihood. For the objection urged in the M&nava Code to Brah- 
mans holding the plough, " that the iron-mouthed pieces of wood not only wound the 
earth, but the creatures dwelling therein 9 ' (X, 84), applies much more forcibly to Jains 
than to Brahmans. The cardinal maxim of the Jain code of morals (and this they 
have steadily adhered to, notwithstanding all changes in the creed itself i, is summed 
up in a single tutra or aphorism, " Ahinsd paramo dharma— regard for animal life is 
the highest virtue." To such an extent is this maxim observed that the more rigid of 
the Yatis, or ascetics of the Jain persuasion, will not drink water till it has been filtered, 
nor inhale air except through a veil, nor set foot on the earth till they have swept the 
spot for fear of unconsciously crashing some unseen insect. Amongst all Jains, 
whether ascetics or men of the world, it is a rule (honoured sometimes in the breaoh 
as well as in the observance) not to light lamps or candles after dark for fear some 
insect may be drawn into the flame, and not to take food after sunset for fear some 
insect may slip between their teeth. The only kinds of occupation fit for a people 
holding such views is to handle rupees or hold a pen ; for in every kind of industry 
but these the destruction of animal or insect life is unavoidable. 

201. In addition to the Jain castes proper there are certain other castes (all of the 
Trading class) which are partly made up of Jains and partly of Hindus, This remark 
applies especially to the Agarw&l, Khattri, Maheshwari, Vishnoi, Jaiswal, and Lohiya 
. castes, and there may be others. As regards the Khattri and following castes, it would 
appear that at the outset they were purely Hindu, but that owing to the greater atten- 
tion which all the trading castes pay to religious matters, and to frequent association 
with members of the Jain castes who follow the same occupation (trade), some of their 
own body have contracted. Jain views of morality and religion. But as regards the 
Agarwil caste, it would appear that the process was reversed, and that all Agarw41s 
were originally Jains, as the Oswals still are. This seems to be implied in the fact, on 
which an Agarw&l, with whom I was once in conversation, laid great stress, that all 
members of his caste, whether Hindu or Jain (and my informant himself was a Hindu), 
abstain altogether from flesh diet. On this ground he claimed that his caste was a 
higher one even than that of Brahman ; and he assured v me that Hindu Agarwals 
were quite as strict on the point of abstaining from meat as their. Jain caste-fellows. I 
have had no opportunity of examining the question any further. There is one fact in 
connection with all these semi-Hindu and semi-Jain fraternities which deserves to be 
specially noticed. Diversity of creed has not in any way impaired the unity of the 
castes, and this shows the groundlessness of the theory, by which caste in one of its 
' aspects has been described as a " spiritual brotherhood/' a body of men" to whom 
a common rite or a common doctrine is everything." Caste, as I have defined it, is 
a functional union bounded by a marriage union ; and now, in verification of this, we 
find two doctrines and two sets of rites co-existing within 'the same caste, and yet 
no disturbance to the marriage union resulting from such diversity. If the bride- 
groom is a Hindu, the girl is married according to Hindu oeremonies and with the 
help of Brahmans ; if he is a Jain, according to Jain ceremonies, in which no Brahmans 
are employed. One of the peculiarities in the Jain marriage rites consists in the cere- 
mony of killing a Brahman in effigy, a survival from the time when Jains and 

Digitized by 


( 120 ) 

Brahmans werel at actual and bittef warfare. A. figure of a Brahman is made of 
kneaded flour, and with hollow trunk, so as to admit of being filled with syrup or 
sugarfed water. The figure do filled is placed under the bench on which the bride 
and bridegroom are seated. After the pair hate been united in wedlock by the Jain 
rifeejthe male relatives of the bridegroom kill the Brahman, catching his blood alias 
the sugared water, as< it comes trickling out of tho cavity of his body, and mixing it - 
•with fresh flour , out of which sweatoneate are at onoe made for distribution to the 
company present In making marriagto contract* no account seems to foe taken 
of scruples which the girl or woman might feel at being mated to a man of different 
persuasion from that in wbkh she was herself born and bred. "What the author of 
the Bombay Census Report has rtoorded concerning the Jains of his own province is 
not less true of the semi-Jain and semi-Hindu castes of Upper India:—" The partition 
between Hindu and Jain is of the very narrowest description,- and cases are not 
uncommon in which intermarriage takes place between the two sections. The bride, 
when she is with her Jain husband, performs the household ceremonies according to 
the»ritual of that form of religion, and on the frequent occasions when she makes a 
temporary sojourn at the paternal abode sho reverts to the rites of her ancestors as 
performed before her marriage. " Between the domestic ceremonies of Jains and 
Hindus 6ne radical difference should be noticed. Jains do not perform the sraddh 
or annual funeral feast for the repose of the dead — a practice the omission of which 
in the eyes of Hindus would be regarded as impious; for the Jain believes that the 
soul of the dead either enters at once into nirvdn y a state of blissful unalterable 
repose far beyond the confines of the earth, or into some new form of earthly exist- 
ence; while Hindus believe that the souk. still wanders in space and suffers hunger 
and thirst, if it is not fed. The mode of worship, too, performed at Jain templet 
differs from that paid at Hindu ones and approximates more nearly to the cultus 
of the Buddhists. There is a frequent use of little bells and candles as at Buddhist 
shrines ; and the idols of the Jinas or deified Saints are not dressed, washed, and fed with 
offerings, as are the images of gods in Hindu temples. The offerings of fruit sometimes 
paid by Jains to their deified saints are not given as food, but as oomplimentary pre- 
sents. Thus in the marriage ceremonies, the funeral rites, and the cultus of idols, . 
there it a considerable basis of diversity between Jains and Hindus, if either section, 
chose to acoentuate the differences : and hence tl\e writer jutft quoted is hardly right 
in saying " that the partition between them is of the narrowest possible description." 
The truth appears to be, not that the differences are inappreciable, but that the 
sentiment of caste, or of " marriage plus functional unionship," has been strong enough 
to hold the two religionists together within the same brotherhood, in spite of the 
diversities that exist. 

202. Among Jains, as amongst Hindus, there are sects of celibate ascetic orders 
who hare renounoed the ties of family and kin and live on the charity of the faithful. 
These are called Yatis, the " restrained or self-masters;" while Hindu monks, as we 
have shown above, are known as Sanny&sis, Atits, or Beshdb&ris. As the census 
report has lumped all kinds of devotees together under the vague name of faqir, 
which, properly speaking, applies to Muhammadans only, it is impossible to say how 
many Jain asoetics there are in the united provinces at the present time. The Jain 
ascetics are distinguished into two great schools — the Svetambar, or those clothed in 
white, and the Digambar, or " those clothed in the four quarters of the horizon, " 
that is, in nothing but air. At the present day Digambars are not allowed to go 
about in a state of nature ; but during their meals, and when two or more are dining in 
company, they revert to the old custom of the brotherhood by discarding their clothes* 
or rags. These men were known to the Greeks, who visited India in about 300 B.C.- 
and later, as the " naked sophists." While Hindu ascetics wear matted locks {j*ta) 
in imitation of their great model Shiva, the Jain ascetics used at one time to pluck out 
their hair, which leads to a mass of short frizzled looks springing up in their place. 
This eiplains why the figures of the Jain saints are represented with frizaled locks.- 

Digitized by 


( 121 ) 

The practioe of depilation, however, is not observed by the Tatis of the present day, 
who either shave their heads or allow the hair to grow long. The Digambars have 
carried the ascetic principle much further than the Svetambars, and are regarded 
in all respects as the older and more orthodox sect. These designations have passed 
from the priests or monks— Yati— to the hearers or laymen, Sr&vak : and thus all 
Jains are expected to enrol themselves under one or other of these two banners. Digam- 
bars approach more nearly to Buddhists, Svetambars to Hindus. The former look to 
Mah&vir, the twenty-fourth of the Jins, as their patron saint and founder ; the latter 
to P4rswa N&th, the twenty -third, who is said to have worn clothes, while Mah&vir wore 
none. The former maintain with Buddhists that women are not capable of attaining 
nirvan or final beatitude, while the latter maintain with Hindus that they are. The 
canonical scriptures of the two sects are not at all the same. It is a remarkable fact, 
as showing the growing convergence of the Jain and Hindu creeds, that the ranks of 
the Yatis are not unfrequently recruited from Hindus of the Brahman caste : and the 
account which Horace Wilson gives of them as "skilful magicians, pretending to skill 
in palmistry and necromancy, dealing in empirical therapeutics and dabbling in 
chemical or rather alchemical manipulations,' 9 implies that they combine the arts of 
the Joshi Brahman, the Ojha Brahman, and the aboriginal wizard or medicine-man. 
A Jain temple is generally managed by some influential layman of the neighbourhood, 
who places some Yati in charge ; and besides these* there is generally a M&li, that is, a 
Hindu of the flower-gardener caste, attached to the service of the shrine. All offerings 
in cash, and the most valuable of the offerings in kind, are appropriated by the 
manager, who keeps the temple in repair ; and a portion of this goes to the maintenanoe 
of the YatL The minor offerings are appropriated by the M&li, who furnishes the 
flower-wreaths and sweeps out the floor of the temple. The function thus assumed 
by the M&li is well in keeping with what we have alluded to before in connection with 
Shivite shrines and offerings (see para. 33). 

203. There is one great ceremony in which Jains do not allow Hindus of any 
kind to participate, and which marks them off more than any other custom that has 
survived as followers of a distinct creed. This is called the Ratha Jatri, and consists 
' in yoking a huge wooden chariot, built two or three storeys high, to a pair of elephants, 
and making them draw it seven times round a large earthen mound, on which idols of 
the Jinas or deified saints have been previously placed for worship. Two instances or 
more of this performance have occurred within recent years in the Lalitpur district, 
and the account which I have received of them was given me by an eye-witness. Some 
wealthy layman, wishing to add to his stock of merit and thereby to increase his chance 
of nirvdn, invites co-religionists from long distances around, even from as far as Gujarat, 
Bombay, or Mysore, and holds a great religions festival which !asts for about one month, 
during which time all the Jains present are fed and maintained at his expense. The 
guests invited bring such idols as they can with them, and these are placed on 
the circular earthen mound or platform prepared for the purpose. On the last day of 
the feast the chariot drawn by elephants is led seven times round the mound, and as 
it passes, cocoanuts, cloves, and various kinds of fruit offerings are thrown at it by the 
multitude seated round it in a ring. If the elephant orushes the cocoanuts with his 
foot as he passes, this is considered fortunate. On the lowest storey of the chariot are 
placed the idols of the man who gives the entertainment ; and in an upper storey is 
seated the man himself with several members of his family, some of whom are dressed 
up as idols. After the ceremony is over, every man in the assembly takes a turban in 
turn (oqe single turban having been prepared for all) and waves it over the head of 
the host The host from this day forth is called a Singhai, the " lion" or chief of Jains, 
and is looked up to with veneration by his co-religionists till the day of his death. The 
use of the elephant in this ceremony for such a purpose as drawing a chariot is unique, 
and the sanctity attached to it as the only animal fit to be used in such a ceremony is 
peculiarly Buddhist. No Hindu deities are allowed to be named on these occasions; 
and the only idota worshipped are those of the Jinas or Tirthankars, the conquerors 


Digitized by 


( 122 ) 

or preoeptora, by the last of whom the religion k said to have been founded more than 
two thousand years ago* 


204. For the Mubammadan castes (if such a thiag as caste can be said to exist 
among persons of this creed) few words will suffice. Caste as an institution is entire- 
ly alien to the Spirit of Isl&m, which has never raised any soruples about the class of 
women who can be taken as wives, or the choice of occupation which its followers may 
adopt The strength of Islim lies in its levelling and democratic character, just as 
that of Brabmanism lay in its conceit and arrogance. Conversion consists in the simple 
process of submitting to circumcision and repeating the kalima or confession : and any 
Hindu on joining the ranks of Islam," not only acquires freedom, but finds a society 
in which he can marry and give in marriage and satisfy the gregarious instincts of 
mankind." Nevertheless it is generally admitted that eaate does exist in a certain 
sense among the Muhammadans of India ; and it will be the object of the remaining 
remarks to endeavour to ascertain to what extent and in what sense, as compared 
with the case of the Hindu castes, this is true. 

205. The Muhammadans of India may be roughly divided into two main 
classes :— (1) those of foreign origin, descended from the original invaders (Persians, 
Afghans, and Turks miscalled Moghals), who crossed the Indus at different times 
and founded provincial or imperial dynasties at the old seats of Hindu power; (2) 
those of indigenous origin, descended) that is, from Hindus or natives of India Hielf 
who were converted to IslAm at different times either by conviction, or through losing 
their own castes as Hindus, or by compulsion on pain of forfeiting their estates or 
in the case of the lower castes, with a view to shaking off the oppression to which thev 
have been hopelessly consigned by the dictates of the Brabmanical creed. The dis- 
tinction between these two classes is not by any means perfect : for Muhammadans of the 
former class, like the Aryans whose invasions preoeded their own by about three 
thousand years, hare undoubtedly taken wives from the people of the land, and there are 
but few of them, as their physiognomy plainly testifies, who can claim to be the paw- 
blooded descendants of the original Muhammadan invaders. There can he scarcely any 
doubt that in the long run the miscegenation will become as complete as it has 
already became in the case uf the Aryans and aboriginals. All race* whieh become 
domiciled in India merge eventually into the type of the native. 

206. Muhammadans of the former class are generally classified under four great 
headings or tribes, ws., Sayyads, Mughals* Piathdns, and Shekhs. Sayyads are chiefly 
of Pfcrso-Arabrc descent and olaim to have inherited, on the male sidp, some portion 
of the blood of the Prophet through his daughter Fatima, who was married to Ali 
the fourth and (as the Shiah sect maintains) the only true Caliph. Almost all Say y*ds 
are Sbiahs as distinct from the other great sect, Sunnie, who pay no special regard to 
Ali "as the Caliph of God," and observe only the tenth day of the Moharram festi- 
val. Mughals and PatMns are invariably, or with extremely rare exceptions, Suonis. 
Shekhs, being themselves of mixed origin, may be either Shias or Sunnis, but are 
mostly the latter. The Mughals * really Turks) are descendants of those MusaWus 
from Central Asia who invaded India under the standards of Timur and Babar. 
The Path&ns are descendants of the Ghaznivites, Ghorians, and other more recent 
comers from Afghanistan. But the name " Path&n " is becoming ambiguous ; for 
it is not unfrequently assumed by Rajput or Chattri converts, who have a partiality 
for this name in consequence of the military reputation associated with it. The 
Shekhs are of no nationality in particular. The name " Shekh " means " venerable " 
and was applied originally to any Musalmins who might be distinguished for 
learning and piety. Men from the Path*n and Mughal tribes have assumed the 
title of Shekh at different times, and, following the example of Hindus, have bequeathed 
the same as an hereditary decoration to their descendants. This, too, is the title 
which Muhammedana of Indian origin (other than Rajputs or Chattris) are in the habit 
of assuming when fhey seek to secure a * higher status in society. Any successful 

Digitized by 


( 123 ) 

tradesman, artisan, or professional man of any kind, if he succeeds in beoomtng 
rich or acquiring social importance in any other way, seeks to dissever the tradi- 
tional connection which binds him to his former Hindu caste by assuming the name 
of Shekh. Moreover, converts from creeds other than Hinduism, such as Christianity 
for example, take the name of Shekh, and are so termed by their co-religionists. 

,207. In no sense can any of these four tribes be called a caste, and nothing but 
confusion as to the meaning of the word " caste " can result from placing them in 
such a category. The three following facts are decisive on this point : — Firstly, they 
can intermarry with each other, whenever individual members happen to be so minded. 
Whatever the tribe of the woman may be, her children take the rank of the father. 
Thus, if a Pathfin or Shekh girl becomes the wife of a Sayyad, her sons are as much 
Sayyads as if their mother had been a Sayyad's daughter on both sides of her parent- 
age. On the other hand, if a Sayyad girl becomes the wife of a Patha-n, her children 
are Pathan and not Sayyad. This is directly opposed to the Hindu principle of caste 
and to the rule laid down by Mann as to the purity of caste blood : — '* In all classes they, 
and they only, who are born in a direct order of wives of tlie same class are to be 
considered as the same in class with their fathers " (X, 5). Secondly, in the event of a 
man's choosing a wife from his own tribe, no attention is paid to the Hindu rule of 
gotra or akardnd (agnatic relationship). Endogamy and not exogamy is the rule among 
Muhammadans of high rank, and this is in direct opposition to Hindu principles. Thirdly, 
there ianot, nor over has been, any pretense tp distinctiveness of function or occupation 
among these four tribes. Any tribe may take np wtth any occupation that individual 
members may select, and there is no siaoh thing as tribal or traditional preference in the 
adaption. Any man. from any of these tribes may aat as Im6m or conductor of prayers in 
the mosque, if his neighbours conaider him to be competent in learning and piety. The 
Sayyad and the Shekh are, U is true, more usually found in this capacity than PathAns 
or Mngbals ; but neither have any monopoly or hereditary rights in the matter, simi- 
lar to 4hc#e enjoyed by Brahmans, Gosbftyens, and Bairigis amongst Hindus. In 
short, the tribe in the Indo-Mnhammadan sense is neither a functional union nor a 
marriage union ; whereas a caste consists in these two unions combined. All that can 
be admitted is that the names by which the above tribes are called denote a certain 
degree of social precedence, and to this extent only can these tribes be said to bear 
any resemblance to Indian castes. For example, a Sayyad is respected by all classes 
of Muhammadans, whatever his personal character or pecuniary status may be, much 
in the same way as a Brahman is respected by Hindus independently of all questions 
of personal merit. Again, more indulgence is shown to the peccadilloes of a Sayyad 
than to those of other Muhammadans : and if a Sayyad is in want, he is much more 
Kkely to be relieved than any other Mnhammadan would be in similar circumstances. 
A Moghal ranks somewhat above a Path&n, and a PathAn above a Shekh. 

208. Although the above tribes cannot be called castes, it is none the less to be 
regretted (as I think) that no statistics have been given in the last census report of . 
the number of persons belonging to each of them, as was done in the separate census 
reports of Oudh and the North- West Provinces published about ten years previously, 
and as has lately been done in the Panjab census report of 1881. We have there- 
fore no data for comparing the changes which may have occurred in the numbers 
durin* the interval of ten years between the last census and the preceding one. 
There°is a Persian proverb current in Upper India, Peshaxin gassdb budem, bddazdn 
qashtem Shekh ; Qhalla chin arzdn shawad, imsdl Sayyad Meshawem—" Last year 
we were butchers, after that we became Shekh ; if the price of food rises, we shall 
this year become Sayyads. - " This shows that even in the case of Sayyad, with 
whom heredity of descent from the Prophet is everything, the question of 
birth is much less keenly scanned than it would be amongst Hindus, and that the 
tribal names are marks of social status as much as of aotual origin. As Mr. Denzil 
Ibbetson has remarked of the Muhammadans of the Panj&b, "the process of 
manufacture in these cases is too notorious for it to be necessary for me to insist on 

Digitized by 


( 124 ) 

it ; and so long as the social position of the Hew claimant is worthy of the descent 
he claims, the true Mughals, Shekhs, and Sayyads, after waiting for a generation or 
so till the absurdity of the story is not too obvious, accept the fiction and admit the 
new brother into their fraternity" (Panjab Ethnography, para. 342). The process, in 
fact, is very analogous to that which I have described respecting the manufacturing 
of Brahmans and Ohattris ; and the Hindu process of manufacture differs from the 
Muhammadan one only in being more gradual and difficult. 

209. Muhammadans of the other class, as distinct from those of the four tribes 
who profess to be of foreign blood, are those of Indian extraction, descended from men 
of Hindu castes who have been converted to Islam. These have preserved to a certain 
extent, as will be explained below, the customs and traditions of the castes from which 
they have sprung. But the information given in the last census report is very 
incomplete as compared with what wan given in the previous Oudh report Of these 
Muhammadan- Hindu castes there are only five whose numbers have been recorded ; 
and the figures (extracted from form VIII. A.) are as follows :— 

1 Muhammadan- Rajputs ... — ... m, ... 122,055 

„ Go jars ... ... *. ... ... 39,85$ 

„ Jfits ... ... ... *• ... 10,401 

„ Tftgts ... ... ... . M ... ff,07O 

„ Mewstis... ... .„ ... ... St,fi66 

Total ... 325,050 

' If we deduct this sum from the total Muhammadan population of the North- 
West and Oudh recorded in the census report, vi$., 5,922,886, there is a remainder 
of 4,677,836 persons of whom no account has been given. This remainder is of course 
made up partly of the four tribes already described and partly of Muhammadan 
converts from the various Hindu castes. The latter are much the most numerous 
body of the two. Even the total of 5,922,886 is obviously incorrect: for several 
Muhammadan castes have been included in form VIIIB., although the castes men- 
tioned in this form profess to be exclusively Hindu (see remarks given above m 
para. 5.) 

210. Muhammadan converts can be conveniently, arranged under two main 
headings : (a) those converted from high-caste Hindus ; (b) those converted from the 
lower oastes. The very fact that such a distinction can still be traced after the lapse 
of so many centuries shows that caste has, in a certain sense, survived the change of 
creed, notwithstanding the utterly casteless character of the creed, to which they 
have declared allegiance. 

211. Regarding Muhammadans descended from high-caste Hindu converts. 
Mr. Williams has written as follows in his Oudh Census Report of 1869, Vol. L, 
page 77:—" The only high Hindu caste among whom there have been occasionally 
large conversions to Muhammadanism is the Rajputs, and the consideration that 
brought it about has in almost all instances in Oudh been the same. Either new lands 
were granted free of revenue or on favourable terms, or lands forfeited by rebellions 
were restored, or some similar privilege or favour was bestowed ; and the Rajput 
tribe forthwith became Mussalm&ns ; that is, they submitted to the rite of circum- 
cision and began to pray in a mosque instead of in a temple, but otherwise they 
retained almost all their Hindu customs and manners." There are a few other high 
castes from which converts have been pccasionally made, vi*. } the Eayasth, the Brah- 
man, and some of the upper trading castes, such as the Ehattri and Bohra. Probably, 
too, many members of the Baidya Or physician caste, now extinct in Upper India, but 
not extinct elsewhere, have embraced Isl&m owing to the higher reputation of Mo- 
hammad an medicine, and taken the name of hakim or Muhammadan doctor. 

212. As regards the Rajput converts, the main points on which they have pre- 
served the tradition of their origin are the following :-— (1) they have retained the 
habit of drinking wine, notwithstanding the prohibition of their adopted creed ; (2) 
in selecting a wife (if they cannot get one from some Mughal, Pathfe or Sayyad 

Digitized by 


( 125 ) 

family, which some prefer to do if they can), they choose one from some Muhamma* 
dan Rajput clan, and never from converts of a Hind a caste lower than their own ; 
(3) they invariably adopt the Hindu rale of exogamy, notwithstanding the custom 
observed in high Muhammadan families of marrying those near of kin ; (4) for nego- 
tiating such marriages they employ a Brahman, who settles all preliminaries as to 
dowry, &c, between the parents of the bride and the bridegroom j (5) in the celebration 
of the marriage ceremonies they employ a Brahman to fix upon the auspicious day for 
each part of the performance and to throw the first handful of straw on the ground 
under the mandap or canopy where the ceremonies are held ; and they give him the 
customary fee, after the Qazi (or Muhammadan officer) has given the seal to the 
union according to Muhammadan rites ; (6) in funeral ceremonies, though they have 
discontinued, the Hindu custom of the sraddha or funeral feast, they still employ a 
Brahman and a Maha- Brahman for certain services to be rendered to the dead imme- 
diately after burial. These customs are, it is true, less strictly observed than they 
used to be ; but the same may be said of all Hindu customs at the present day. even- 
amongst those Hindus who have not adopted Isl&m as their creed. 

The same remarks apply to the Muhammadan Kayasth ; but the conversion of 
Kayasths to Isl6m has of course arisen out of different motives. 

213. The number of Muhammadan Brahmans is much smaller, and their descent 
from the Hindu priestly caste cannot be so easily traced. Mr. Denzil Ibbetson, in 
his outlines of Punjab Ethnography (which consists of extracts from his late cen- 
sus report), alludes to them in the following terms in para. 512 : — " There are 3,500 
MussalmAn Brahmans, chiefly in the Delhi district. These men are known as 
Huseni Brahmans, and are said to receive oblations in the name of the Hindu 
gods from Hindus and in the name of Allah from Mussalm&ns." I have heard of the 
same kind of Brahmans in these provinces also. There is another kind of semi- 
Muhammadan and semi-Brahman priest ; and this is to be found at the tombs of 
saints, at which Hindus and Muhammadans alike go to pay their offerings. For 
example, there is the shrine at Kachauohha, in the Fyzabad district, where Makhdum 
Sahib, a Muhammadan saint, was buried ; but it is well known that this was once a 
Hindu temple presided over by Kamal Pandit, whose soul is still worshipped there in 
common with that of the Mussalmdn Makhdum. Invalids' of either sex go there to 
be cured of blindness, lameness, infecundity, and the various diseases ascribed, by 
Hindus and Muhammadans alike, to demoniacal possession. The priests wbo are 
in charge of this shrine and who receive the offerings on behalf of the saints are 
said to be descended from those Brahmans who were owners of the temple before it 
became the shrine of a Muhammadan saint. Other instances might be named. Such 
men constitute a kind of caste, for their office is hereditary and they take presents 
from Hindus as well as Muhammadans. Both of these peculiarities favour the 
supposition of their Brahmanical origin, for there is no such thing as an hereditary 
priesthood amongst Muhammadans ; and the custom of feeding or feeing temple 
priests, in the hope of propitiating the god or saint who resides at the shrine, is 
peculiarly Hindu. 

214. Muhammadans descended from lower-caste Hindu converts are very 
numerous. In addition to the few castes named in the census report (see para. 209) 
we may enumerate the following. Wherever it is possible, I have given in a bracket 
the Muhammadan name against the corresponding Hindu one. In one case (that of 
Darzi), the Muhammadan name " Darzi " has already entirely superseded the Hindi 
name " Suji" for tailor ; and in another instance (that of confectioner,) the Muham- 
madan name " Halwai" has superseded the Hindi name Mithiyfi. In two cases, mi., 
liangrez and Dom Mirfisi, no Hindi name exists. Both o f these have been marked 
with an asterisk : 

Digitized by 


( 126 ) 

I.— Bhangi (Kihtar, Lai Begi) 

f Dhobi (Gazur) ... 

J Kunjra/Mevirafarosh, Sabaifarosh) 
L j GadduGbosi) 

[ Khatik (Cbik or Chikwa) 

Suji (Darzi) ... ., 

Nai (Hajjam) 

Kahar (Bhisti) .♦, 

Bunkar or Kori (Jul&h&) 

Rangrez* ... 



Dhuniya (Naddif) 
Kurphdr (Kasg4r) 

Barhai (Najj&r) 

Mithiya (Halwai) 
Dom Mirasi* 














Oloth -printer. 






215. The industries of the above fraternities have been briefly described already 
in connection with their Hindu equivalents, and what we have now to do is to attempt 
to show to what extent these fraternities can be called castes. C&ste,as/we have seen, 
implies three things : (a) heredity of function, (h) prohibition against marrying with out- 
siders, (c) prohibition against eating and drinking with outsiders^ The Mubammadaa 
fraternities certainly possess the first ; for heredity of function is of the very essence of 
their being. As a general rule they possess the secopd also ; for though intei-marriage 
with other fraternities is not prohibited by rule, on pain of expulsion, as amongst 
Hindus, yet it is discouraged by public opinioa and is seldom or never practised 
iu fact. The third condition is not observed in the case of any of the above fraterni- 
ties except the first, that of the Lai Begi or sweeper, which in all respects is as 
complete a caste as any Hindu caste could be. The headings 1, 11, III are intended 
to signify the order of social precedence in which the several fraternities stand 
to each other, and here again, something of the caste system of the Hindus cornea ont. 
Thpse under heading III are the classes or castes from whom the Shekh tribe is 
occasionally recruited^ independently of its natural increase from within. 

216. Roughly, then, it may be said of the above fraternities that they are caste* 
of the Hindu pattern,, so. far as the two principles of trade-unionism and marriage 
unionism are concerned, but not (except in case of the sweeper) so far as convivial 
intercourse is concerned. That they consider themselves to be castes like their 
Hindu brethren is certain; for Mr, Williams remarks in the Oudh Census Report,, 
Vol. I, page 81, that " there were separate columns in the enumerator's return 
for caste and profession, and the instances were simply innumerable in which 
entries were made h* both columns for persons with Muhsmmadan names; " and this 
is confirmed by the experience of every one who has enquired ipte the subject. If a 
man tells you that he is a Hajjam by caste, and you find that he has been earning bis 
livelihood from a boy as a gardener, it is evident that he looks upon caste from the 
same point of view that a Hindu does. The case here described is not an imaginary 
one, and similar ones are of constant occurrence. Most of the above fraternities,. 
too> have retained the Hindu caste custom of settling disputes by a panchaytt or 
council of arbitrators selected from among the brotherhood. If the subject brought 
before the panch&yet relates to some question of trade or business affecting the fraterni- 
ty as a whole, Muhamadans will be found associating with Hindus and forming part 
of the same assembly. Moreover, all the Muhammadan castes or trade-unions observe 
the Hindu rule of not taking a wife from within the ghardna. or circle of agnatic relation- 
ship reaching back to the fourth generation. If we had space to discuss the customs* 

Digitized by 


( 1*7 > 

and traditions, of each Muftammadan caste individually, it would be found that each of 
them retains some link connecting it with ft* Hindu anqestors. For example the Chik 
or Chikwa caste of butcher is the Muhammadan counterpart" of the Hindu Khatik, But 
the Chikwa has so far retained the tradition* of hi* descent that he whTnoislky the ox r 
sp that a new class of Muhammadan. butcher, called by the generic- name of Qas6b, 
has come into existence, in order to supply Muhammadans with their favourite flesh diefy 
beef.. The Chikwa will not intermarry with a Qasdb, but ooly with Chikwas;and thus 
the Muhammadan fraternities come within the description of " a functional union 
bounded by a marriage union/ 9 which. I hare given, in |para. 192 as the minimum. 

definition of caste. 


217. In further proof of the fact that the lower classes of Muhammadans are 
castes in the Hindu sense of the word, Mr. Williams adds that " they consider the 
marriage of widows unlawful" (page 79). This, however, is not merely irrelevant ar 
an argument, bnt is incorrect inpoint of fact* Every one of the Muhammadan castes- 
enumerated in para*/214 allows the practice of widow remarriage, and so does every 
corresponding. Hindu caste. It is a popular error to suppose that Hindus do not 
allow their widows to re-marry. The prohibition is quite of modern* date, and k even, 
now confined to- the few " twice-born" oastes who stand at the top of the tree. Among 
all the middle and lower castes, Hindu and Mnbammadan* alike, widows can be and 
are Demanded without reproach. High-class Muhammadans have in some instances- 
imbibed a prejudice against the custom from high-caste Hindnsr but this is entirely 
opposed to the teaphing of the Qur&n, whicLstr fatly disallows widows to remain single. 

218. It is nutoh to be regretted (as T think) that xHvdetaHs have been given in 
the last census report as to the number of Mnhammudans represented in each of the 
several castes. A, comparison of such figures with those given in the previous census 
of OudS and the North- West Provinces would have shown what progress Muham- 
madanism is making against the indigenous creed of Btahmanism. Mr. Williams' 
resaarks on this point are worth quoting r— a It appears to be an almost invariable rale 
that in* those professions which Muhammadans have adopted they form. the great majo- 
rity, and probably they do in time succeed in monopolizing every trade they adopt; 
By this 1 do not mean that their competition drives out the Hindus (for Muhammadans 
are generally not so industrious as- Hindus), but that all the Hindus of any caste, the 
principal occupation- of which, has been, invaded by Mnssalm&ns, eventually, become 
Mussalm&ns. The conversion is rendered easier when it does not involve the- aban- 
donment of an hereditary trade ; * ** •'and when this is not a necessary, accompa- 
niment of a change- of religion, the attractions of comparative* freedom from caste 
rules,, greater laxity in the choice of food, less subjection to^ priestly influence, 
and a certain rise in* the social scale, all of which Muhammadanism* offers to the 
new convert, become too great* to be resisted. The Saikalgars (cutlers^ metal- 
polishers, or sharpener*) are an instance. They are now exclusively Muham- 
madan ;. in Eae Bareli alone, the district where Muhammadans are fewest, there 
are a few Hindus still following this profession. The name of the (Hfndu) 
caste is Baria; elsewhere they all seem, to have become Muhammadans and call 
themselves Saikalgars." Many other instances might be quoted of Hindu castes 
becoming almost, if not entirely, Muhammadan^ To quote from the Panjab census 
report lately compiled by Mr. Denaul Ibbetson (and the same proportions are likely 
to apply without much difference to the North- West and Oudb,. if we make due allow- 
ance for the fact that here the rate of Hindus to Muhammadans is about fifteen per 
cent, against about fifty per cent in the Panjab), in the Mochi or cobbler caste there are 
only 1,366- Hindus, bnt 347,900 Muhammadans ; in the Dhobi or washerman caste 
there are 23,317 Hindus, bnt 109,889 Muhammadans; in the Rangrez or dyer caste 
there are 103 Hindus, but 4,957 Muhammadans ; in the Kunjra or green-grocer caste 
there are 72 Hindus, but 5,928 Muhammadans ; in the Kam&ngar or bowmaker and 
painter easte there are 264 Hindus, but 2,89ft Muhammadans ^ in the Manihir or 

Digitized by 


( 128 ) 

bracelet-maker caste there are 2,950 Hindus, bat 57,739 Muhammadans ; in the Darsi 
or tailor caste there are 10,022 Hindus, but 22,441 Muhammadans. It is evident that 
the Hindu element in some of these castes must soon cease to exist ; for isolation and 
the difficulty of procuring wives will soon rettder their position untenable, and Hinduism 
has nothing to offer them in compensation for these inconveniences. The above 
examples are taken exclusively from the Artisan and serving castes, who dwell chiefly in 
towns. But there is reason to think that 'a similar revolt against Hinduism is 
commencing in villages also among the rural or agricultural population. But as 
the figures are wanting, this is only matter of conjecture. 

219. As amongst Hindus and Jains, so also amongst Muhammadans, there are 
several sects of religious orders, which differ, however, from those of Hindus and 
Jains in the fact that marriage is an openly recognized practice. The place where 
a family of devotees resides is called a takiah, that is, a " pillow " or resting-place 
from the cares of the world ; and the dwelling-house may be either a hovel made of 
mud or straw Use a hermit's hut, or a well-built masonry mansion. The paterfamilias 
in such establishments leads a perfectly idle life, beyond reading or repeating the 
Kuran and receving what is given to hirta by the pious. Some are quite illiterate, 
but are supposed to make up for their want of reading by thiuking very earnestly 
and profoundly. As amongst Hindus, the number of these Muhammadan 6ects is 
very great and, like Hindus, they have two great stages of asceticism. The first 
and easier is called that of the S&lik, "one who treads in the appointed path," and 
this is the counterpart of the Hindu banaprastha. The second and most severe, 
being the final one, is called that of the " Majzub," that isj " one who is absorbed in 
the divine nature": this is the counterpart of the Hindu Sannydsi (see para. 138). 
This idea of absorption into the divine spirit is not at all in character with the per- 
sonal beatitude promised in Muhammad's Paradise, and has evidently been borrowed 
from Brahmanical teaching. For reasons already stated in paras. 150 and 202 it is 
impossible to ascertain from the statistics shown in the census report how many of 
the devotees recorded under the name Faqir are Hindu, hew many are Jain, and how 
many Muhammadan. 

220. I would beg to offer one or two suggestions in cdnclusion as to the form 
in which the statistics of the next census might be compiled. Against each caste 
name there should be four columns, headed " Hiudu, Jain, Muhammadan, and 
Others," so as to show the number; of persons within each caste who belong to each 
of the three main creeds. This is the method which has been followed by Mr. Denzil 
Ibbetson in compiling the results of the last Panjab census. Something similar to 
this was done in the Oudh census report of 1869 compiled by Mr. Williams. But 
(as we have already remarked) in the last census report of the North-West and Oudh 
there is nothing to show to what extent the different religions are represented within 
the several castes. 

221. I would also suggest that against each caste name there should be two 
more columns, one showing the original fnnction of the caste — that from which its 
specific character and, in almost all cases too, its name have been derived — and the 
other showing the various other occupations besides the original one followed at the 
present time. If these suggestions are accepted the form of the statement would be 
as follows :-*- 

Name of caste. 
















C 1 


Numbers by religion. 






Numbers by set. 



Digitized by 


( 129 ) 

222. I have already said something about the synonyms of caste names in the 
places where 1 have described each group of castes. Bat in page xxxv, Appendix 
No. 12, of the census report, a list is given of the " local names of certain castes and 
of the caste in which they have been included in form VI11." Some of these iden- 
tifications, which have not yet been alluded to, appear to be incorreot. Thus Gangd- 
putra, Gangab&si, and Ghitwil are set down as synonyms or local names of Bh4t • 
whereas they undoubtedly belong to the general head of Brahman and are never 
applied to the caste of Bh4t. Again, Bhagat is set down as a local synonym of 
Cham&r ; whereas Bhagats exist in every caste and not merely in that of Cham&r ; 
for the name is applied to those individuals of any caste who make a merit of abstain- 
ing from flesh and wine. Again, Kathak is set down as a synonym of Kanchan, 
B6dha, and Ramjani : but the caste is as distinct as possible ; for the Kathak is a high 
and much respected caste of musicians who wear the Brahmanical thread ; while the 
others are hereditary prostitutes, who combine this vile function with that of singing 
and dancing. Again the Beldar is set down as a synonym of Ronii ; but the former 
is a general name for navvy and is not even a caste, while the latter is one of the 
trading castes* 

223. As it is presumed that these identifications have been used in compiling 
the figures in form VIII, the number of persons entered against the castes correspond- 
ing cannot be correct. And there are certain other entries, not previously alluded 
to which appear to be inaccurate also. For example, against the Hindu caste of 
Manihar there is an entry of only 6,612 persons of both sexes. But if we assume 
that there is an equal number of Muhammadans belonging to the same caste, even 
then a total of 13,000 persons seems a very small number for a caste whose function 
of making glass bangles is almost universally employed by the female half of a popu- 
lation of about 44,000,000 souls (see para. 60). Again, there is an entry of only 
1,385 persons of both sexes against the Brahmanical sub-caste of Mah&-Brahman or 
" funeral-priest." Considering that his functions as such are indispensable to a popula- 
tion of over 38,000,000 Hindus, and that individual families of this caste are widely 
scattered among the towns and villages of the united provinces, the number of 1,385 
must be a great deal smaller than the fact. Again, against " Beng&li " (one of the 
foreign races ftot returned by caste") there is an entry of only 2,521 persons of both 
sexes. But it is well known that in Benares alone there are more than three times 
as many as this number, and that in all the chief stations of these provinces the 
number of Bengalis has been steadily increasing for several years past. 


28*A February, 1885. Inspector of Schools, Oudh. 

Gboup I. 

—Casteless Tribe*. 

Serial No. 

Name of tribe. 

Census of 









6 664 


Kan jar 





















• •• 



Bar war 



• •« 












• •• 


Mixed op in censas with Babelija. 













Mixed up in census with Ban- 







na anush. 












































Digitized by 


( 180 ) 
Group IIA.— Castes allied to Hunting State. 

Serial No. 

Name of caste. 

Census of 1 



■•• ••• 




••• ••• 





••• ••• 




A rakh 

••• IH 


64 ,7 13 








MM ,,» 





••• ••• 





• •• ••• 





... ••• 





... ... 





•— • •• 




not returned by caste 
Gbano Total 

• •• 





Omitted in census. 

Tarikasb, &c 

Group I IB. — Costes allied to Fishing State. 

Serial No. 

Name of caste. 








































Occupation not returned by 



Gband Total 


Census of 








• 299 



(Nos. 6, 7, 8, and 9 are not 
named ic census, but are in- 
, eluded under the genera) name 
j of Uahar with 1,222,672 per- 
' sons. 


Group IIC — Castes allied to the Pastoral State. 

Serial No. 

Name of caste. 

Census of 



Goshi or Gaddi ... 

A har ,., M# ,,, 

Gujar ... , M 

Gatlariya ... 


AUir ••* ... *.. 

J»t ** ... „, 

67 *,068 




Group II D. — Agricultural Castes. 

Serial No. 

Name of caste. 

Census of 




••• ... 


3,000 599 



... ••• 





... ••• 





... ... 





••• ... 





■•• ••• 





•*• ••• 


2,2< 5 892 



••• ..« 


20V 7 7 



•*• ... 





..% •»« 


10) 6J5 



••• ... 


• «• 




Gband Total 



Perhaps intended for pattidar. 

, 6,004,067 

Digitized by 


( 131 ) 
GROUP HE. — Landlord and Warrior Caste. 

8erial No. 

Name of caste. 

Census of 



Cbattri or Rajput 


Group IIIA. — Artisan Castes preceding the stage of Metallurgy. 


Name of caste. 

Description of arts. 

Census of 



Bansphor or Dbarkar 



Mochi m 


Kori and Julaha ... 



Kambar ... 


Basket-maker ... 

Leafpiate-maker „. 

Tanner ... 


Thread -twister 

Wearer ... 

Oil- maker „, 

Spirit distiller ... 


Salt-ex trader 












Name of Mochi not 
gi?en in census j 


Group 11 IB. 

— Artisan Castes coeval with Metallurgy. 


Name of caste. 

Description of handicraft. 

Census of 




Stone-cutter ... 



Kamangar and Tirgar 

Bv)w-maker and painter ^ 





Carpenter ••• 





Wood -carver M , 





Iron smith ... 





Bra as moulder M 





Brass engraver ... 

27,3 1 a 




Goldsmith ... 





Bracelet-maker m* 















Braid-maker ... 










Dyer „ 





Confectioner ... 
Total .* 



Group IV.— Trading Castes. 


Name of caste. 

Description of trade. 

Census of 





Forest-trader ... 





Green grocer ... 





Grain-parcher ... 










| Pedlar a and small retail 





y dealers who seldom keep 





I regular shops. 







Not named in census. 












• *• 

1 Traders who keep regular 
! shops aud deal in larger 
f quantities of grain, apices, 
I perfumes, cloth, &c. 


















• •• 

! Bankers, money-lenders, 
j wholesale dealers, &c 










Traders not returned by 


Called in census 



Baniya andGandbi. 

1,67 1,390 

Digitized by 


( 132 ) 
Group V.— Serving Castes. 


Name, of caste. 

Description of service or 

Census of 





Bhangi ... 



Napit or N&i %.. 

P&wariya .♦. 


Doiii Miraei M . 






Water-carrier and general 

Barber and snrgeon 

£ Singers and mnssicians ... 

Bard and genealogist ... 
Estate manager and writer... 

Total -, 


618,872 ' 






Not separately named 
in cenaus. 

Ditto. ' 


Group VI. — Priestly Castes. 


Name of caste, 

Description of function. 

Census of 




Brabmans in general 
Special Brahmans— • 

(a) Ojha 

(b) Panda 

(e) Gangiputra 

(d) Joshi ... 

(e) Mab&-Brahman 

Miscellaneous functions ... 









Not named in census. 

Including Ach&raj. 


Group VII. — Religious Orders. 


Name of order. 



Census of 




Foreign Races not returned by Caste and " Unspecified." 


Name of ra 


. 1 






8 - 








• •• 















I — 





Digitized by 



Digitized by V^OOQlC 

Digitized by 


Digitized by 


Digitized by 


Digitized by 


JUN 1 3 1939 

Digitized by